T 1491/21 27-09-2024

Reasons for the Decision

1. Admissibility of the appeal (Article 108 EPC)

1.1 In the statement of grounds of appeal, the appellant should indicate the reasons for setting aside the decision under appeal, or the extent to which it is to be amended, and the facts and evidence on which the appeal is based (Article 108, third sentence, EPC and Rule 99(2) EPC).

1.2 Respondent O1 submitted that the appellant's reference to earlier submissions at first instance in its "ANNEX 1 - Submissions in the event of non-remittal" could not be considered sufficient for the purposes of Article 108, third sentence, EPC (see the last paragraph on page 3 of respondent O1's reply of 22 March 2022).

1.3 The decision under appeal is limited to the issues of added subject-matter and sufficiency of disclosure. Respondent O1 has not contested that the appellant has indicated the reasons for setting aside the decision under appeal by addressing these issues in its statement of grounds of appeal.

1.4 The appellant's annex (starting on page 13 of its statement of grounds of appeal) relates to issues which the opposition division had not decided upon. It refers to documents submitted in the first-instance proceedings before the opposition division. From a formal point of view, this approach does not contravene the requirements of Article 12(3)(a) RPBA, under which documents referred to in the statement of grounds of appeal do not need to be annexed if they were filed in the course of the opposition proceedings. On the substance of the submissions, the appellant cannot be required to modify or supplement its submissions on a matter which has not been decided on by the opposition division in order for its appeal to be admissible.

1.5 In view of the above, the statement of grounds of appeal complies with the requirements of Article 108, third sentence, EPC, and therefore the appeal is admissible.

2. Patent as granted - added subject-matter (Article 100(c) EPC)

2.1 The ground for opposition under Article 100(c) EPC prejudices the maintenance of the European patent as granted if its content extends beyond that of the application as filed.

A patent application or patent can only be amended within the limits of what a skilled person would derive directly and unambiguously, using common general knowledge, and seen objectively and relative to the date of filing, from the whole disclosure of the description, claims and drawings of the application as filed. This is often referred to as the "gold standard". After the amendment the skilled person may not be presented with new technical information (see Case Law of the Boards of Appeal of the European Patent Office, 10th edition, 2022, "Case Law", II.E.1.1).

2.2 In points 9.1 and 9.4 of the decision under appeal, the opposition division concluded that the amendments made to feature 1.4 satisfied the requirements of Article 123(2) EPC because they were directly and unambiguously derivable from the wording of the corresponding feature in the application as filed.

2.3 Feature 1.4 of claim 1 as granted specifies the following:

"identifying plausible combinations, said plausible combinations including a light sheet label of said projected pattern and the respective unique identifier for two plausible blobs where each plausible blob is selected from said blobs in each of said images respectively, said plausible blobs crossing the conjugate epipolar line of the respective image;"

2.4 The corresponding feature 1.4 in claim 1 as filed reads as follows:

"identifying plausible combinations, said plausible combinations including a light sheet label of said projected pattern and said unique identifier for a plausible blob selected from said blobs in each of said images, said plausible blob crossing the conjugate epipolar line;"

2.5 Respondent O1 argued that the corresponding feature as filed disclosed two items in the plausible combinations and could not provide a basis for the amendment to feature 1.4 of claim 1 as granted. The opposition division had relied on the disclosure of paragraphs [0046] and [0050] of the application as published (see section III. of respondent O1's reply of 22 March 2022). However, paragraph [0050] disclosed that the triplets, i.e. the combinations, used curves instead of blobs. Contrary to what was argued by the opposition division, blobs and curves could not be used interchangeably since paragraph [0046] distinguished between them. Moreover, paragraph [0046] did not disclose a plausible combination comprising two blobs and a light sheet label.

Furthermore, the expression "in each of said images" could not refer to the identifiers of the blobs because they were not in the images. The expression had to be interpreted as referring to the blobs in the images.

In addition, if a feature of a claim as filed required interpretation, it could not provide a direct and unambiguous disclosure for an amendment. Interpreting the features of a claim as filed to determine its disclosure was not reconcilable with the principles of the gold standard.

2.6 Respondent O2 submitted that claim 1 as originally filed defined a duet composed of a plausible blob selected from the blobs of the two images and one light sheet label (see section II. of its reply of 21 March 2022). This was a technically sensible interpretation, in particular in view of paragraphs [0003] and [0004] of the application as filed. The amendments to claim 1, specifying that the combination included (two) blobs, redefining the plausible combination and stating that each of the plausible blobs in each of the respective images crossed the conjugate epipolar line of the respective image, were not derivable from claim 1 as originally filed.

According to respondent O2, paragraph [0050] of the application as filed did not provide a basis for feature 1.4 of claim 1 of the patent as granted. This paragraph did not mention blobs and disclosed that a reduced number of combinations were determined by selecting plausible combinations on the basis of a matching error (while claim 1 used the matching error after the selection) and comparing the matching error of each of the possible combinations with a threshold. Therefore, paragraph [0050] disclosed a different embodiment. Moreover, this paragraph required a light projector unit projecting the pattern and two cameras of the sensor to be calibrated in the same coordinate system. Extracting an isolated feature from the specific embodiment disclosed in paragraph [0050] amounted to an intermediate generalisation.

2.7 The appellant concurred with the opposition division. It also submitted that selecting two plausible blobs was unambiguously clear from originally filed claim 1 as such. This was confirmed when reading the claim in combination with the description as filed, in particular paragraph [0050] (see also section 2.1 of the statement of grounds of appeal). Since the basis for feature 1.4 of claim 1 as granted was claim 1 as originally filed, there was no intermediate generalisation resulting from omitting features disclosed in paragraph [0050].

Paragraph [0004] of the application as filed disclosed using points of a single image but was part of the description of the background art and should not be used to interpret the claims. The only technically sensible interpretation of feature 1.4 as filed in the context of claim 1, which specified "matching points between two images of a scene", was that a plausible combination included an identifier of a light sheet and two blobs, one in each of the two images.

2.8 The board agrees with the appellant.

2.9 Claim 1 as filed specifies plausible combinations including "said unique identifier ... in each of said images". In the board's view, this phrase, in the context of the application, directly and unambiguously defines a unique identifier in each of the two images specified in claim 1, and the plausible combinations thus include two of said unique identifiers, one in each of the images.

2.9.1 The board is of the view that to determine the disclosure of a claim, its subject-matter must first be determined by interpreting the claim (i.e. establishing the meaning of the claimed features) from the perspective of the person skilled in the art. The meaning of claimed features is determined by what the person skilled in the art would understand when reading the claim, using common general knowledge, and seen objectively and relative to the date of filing, in the context of the application as filed.

2.9.2 The board is not convinced by the arguments of respondents O1 and O2 that the person skilled in the art would infer from the wording "said unique identifier for a plausible blob selected from said blobs in each of said images" that only one unique identifier was included. The board agrees with the appellant that it follows from the natural reading of the wording that there are as many identifiers as there are images.

2.9.3 The board accepts that, under certain conditions, a point in one image and a light ray may suffice for calculating the coordinates of a point in three-dimensional space. This is disclosed in paragraph [0004] of the application as filed as a prior-art technique which exploits a known orientation of a projected ray and its reflection on a single image. It is not apparent, though, how the single image of the prior art could serve the claimed purpose of "matching points between two images".

2.9.4 However, the board is not convinced that a claim should be interpreted according to a specific disclosure of the background art. The application as a whole consistently teaches matching points between two images using triplets (i.e. a light sheet label and two identifiers for two blobs, one in each of the two images) and matching errors of the triplets (see e.g. paragraphs [0050] to [0054] of the description as filed). It is also apparent that the identifiers (indexes, labels) for blobs and curves in paragraphs [0046] and [0050] and feature 1.4 of claim 1 as filed refer to the same entities.

2.9.5 Therefore, the board considers that the person skilled in the art would construe the plausible combinations of feature 1.4 of claim 1 as filed solely to mean triplets, this also being the only technically sensible interpretation in the context of the claim. Therefore, claim 1 of the application as filed directly and unambiguously disclosed combinations of triplets.

2.9.6 The argument that the amendment represents an intermediate generalisation is not persuasive either. The board agrees with the appellant that since the basis for the amendment to feature 1.4 of claim 1 as granted is claim 1 as filed, the context of the disclosure of triplets in paragraph [0050] is inconsequential for assessing whether the subject-matter of claim 1 as granted extends beyond the content of the application as filed.

2.9.7 The amendments to claim 1 as granted specify that the conjugate epipolar line crossed by a plausible blob is that "of the [blob's] respective image".

Claim 1 as filed defines extracting blobs from two images and "a conjugate epipolar line on each of said images". The plausible blobs cross the conjugate epipolar line. Since the blobs and epipolar lines are confined to a particular image, the wording of claim 1 as filed implies that a plausible blob crosses a conjugate epipolar line on the image from which it has been extracted.

2.10 In view of the above, the subject-matter of claim 1 as granted does not extend beyond the content of the application as filed. Thus, the ground for opposition under Article 100(c) EPC does not prejudice the maintenance of the patent as granted.

3. Patent as granted - interpretation of claim 1

3.1 In the case in hand, the board deems it necessary to interpret claim 1 as a precursor to the assessment of sufficiency of disclosure.

3.2 The boards of appeal have laid down and applied the principle whereby the description and drawings are used to interpret the claims and identify their subject-matter if the claim language is ambiguous (see Case Law, II.A.6.3.1 and II.A.6.3.3).

3.3 The interpretation of the phrase "plausible combinations" in feature 1.4

3.4 The appellant submitted that the composition of the combinations and the requirement that the blobs crossed epipolar lines specified in feature 1.4 provided a clear definition of the phrase "plausible combinations". Feature 1.4 was thus clear in itself and there was no need to interpret it in light of the description. However, if the description was to be considered, paragraph [0033] of the patent specification identified which blobs were considered for the combinations, and paragraph [0036] disclosed that plausible combinations already existed before the number of them was optionally reduced using error measurements.

3.5 Respondent O2 argued that the person skilled in the art would not have seriously considered the plausible combinations to be those whose blobs crossed the conjugate epipolar lines (see section III.1 of respondent O2's reply of 21 March 2022). The description did not allow for any such interpretation; only the claims did. Paragraph [0033] of the patent specification disclosed blobs crossing the epipolar line but did not define what "plausible combinations" meant.

Therefore, according to respondent O2, there were two possible interpretations of the phrase "plausible combinations":

(a) It included the additional requirement to calculate matching errors for all possible combinations before identifying "plausible combinations"; this interpretation was consistent with the teaching of paragraph [0035] of the patent specification.

(b) It was the definition submitted by the appellant.

3.6 The board is of the view that feature 1.4 of claim 1 as granted provides a clear, unambiguous and technically sensible definition of the plausible combinations.

According to feature 1.4, plausible combinations include an identifier for a plausible blob in each of two images. Feature 1.4 defines a plausible blob as a blob which crosses the conjugate epipolar line of the respective image. The plausible combinations include these specific blobs, not just any blob. Hence, on a normal reading of the claim, the person skilled in the art would have understood that the plausible combinations include those blobs which cross the conjugate epipolar line of the respective images.

3.7 Moreover, since the definition of the phrase "plausible combinations" leaves no doubt as to what is meant, the issue of whether the description provides a corresponding disclosure may be disregarded when interpreting feature 1.4. Respondent O2's line of argument that the "plausible combinations" included additional requirements not claimed (see point 3.5(a) above) appears rather to amount to an objection of lack of support under Article 84 EPC. However, the requirements of Article 84 EPC are not applicable to the claims of the patent as granted (see G 3/14, OJ EPO 2015, A102, point 55 of the Reasons).

3.8 In view of points 3.4 to 3.7 above, the board considers that the person skilled in the art would construe the phrase "plausible combinations" to mean those combinations having the characteristics specified in feature 1.4 of the claim.

3.9 The interpretation of feature 1.7

3.10 During the oral proceedings, respondent O2 argued that feature 1.7 was linguistically unclear and gave rise to multiple technically sensible interpretations.

3.10.1 If the phrase "for each epipolar plane" referred to the figure of merit, then a figure of merit would be computed for each epipolar plane. If it referred to the matching errors, then a figure of merit would be computed across the epipolar planes. If it instead referred to the most probable combination, then there would be a separate most probable combination for each of the epipolar planes.

3.10.2 The wording of feature 1.7 cast doubt on whether a separate figure of merit was calculated for each of the plausible combinations.

3.10.3 In summary, from a purely linguistic view the ambiguities in the claims allowed for two possible interpretations:

(a) A separate most probable combination was determined for each epipolar plane by computing a single figure of merit using the matching errors of the plausible combinations in the corresponding epipolar plane or by equating each figure of merit for each plausible combination with the matching error of the corresponding plausible combination.

(b) A most probable combination was determined by computing a figure of merit using matching errors for all epipolar planes.

3.10.4 The description and figures of the patent did not support the interpretation that a figure of merit was computed using matching errors for all epipolar planes. In particular, Figure 8 did not support this interpretation as it showed that the figure of merit was determined across some epipolar planes but not over all epipolar planes as specified in claim 1. Instead, Figure 9 explicitly showed that a separate figure of merit was calculated for each epipolar plane and implicitly disclosed calculating a matching error.

3.10.5 In addition, interpreting feature 1.7 as specifying that a single most probable combination was determined was technically nonsensical. It followed from basic physical considerations that the same number of most probable combinations as light sheets had to be determined, one for each of the light sheets. However, paragraph [0038] of the patent specification disclosed rejecting matches if more than one triplet presented a low average error. Therefore, the description did not support an interpretation according to which the same number of most probable combinations as light sheets were generated.

3.11 The appellant argued that feature 1.7 should be interpreted in the context of claim 1, which defined a loop over epipolar planes to generate one matching error per plausible combination and epipolar plane. These errors were then taken into account to determine a figure of merit and a most probable combination. Paragraph [0038] of the patent specification disclosed that for each plausible combination a figure of merit was computed using the corresponding matching errors which had been calculated for all epipolar planes. Figure 8 plotted the matching error against the epipolar index, and a figure of merit was determined for each of the curves in the graph.

The person skilled in the art would not consider an interpretation in which the figure of merit used a matching error for a single epipolar plane. This would render determining the matching error redundant because it would be the same as the figure of merit.

Similarly, the person skilled in the art would reject an interpretation in which only one overall figure of merit was computed using matching errors from all plausible combinations. In that case, it would not be possible to distinguish between the plausible combinations and find a most probable combination.

The person skilled in the art would readily realise that a matching error might not be available for all epipolar planes. The description of the patent in suit acknowledged that not all blobs crossed an epipolar line, meaning that the calculation of the figure of merit could not take into account epipolar planes for which a matching error had not been calculated. Moreover, feature 1.7 referred to the matching error computed for plausible combinations, i.e. those whose blobs crossed the epipolar line.

Figure 9 and paragraph [0040] of the patent specification could not be regarded as a basis for interpreting feature 1.7 because they did not disclose calculating the matching errors specified in claim 1.

3.12 The board finds that the only contextually correct and technically sensible interpretation of feature 1.7 is that a most probable combination is determined by computing a respective figure of merit for each of the plausible combinations, each figure of merit being computed using the corresponding matching errors for all epipolar planes. The reasons are the following.

3.12.1 The diverging interpretations put forward by the parties partly result from uncertainty a priori as to whether the phrase "for each epipolar plane" refers to "a figure of merit", "a most probable combination" or "said matching error".

3.12.2 However, features 1.3 to 1.6 of claim 1 specify a loop in which, for each epipolar plane, a plurality of matching errors are calculated. Thus, the expression "said matching error" does not have a defined antecedent and must be read together with "for each epipolar plane". It follows that interpretations in which the figure of merit is computed for each epipolar plane are contextually incorrect. A similar conclusion would apply if the phrase referred to a most probable combination.

3.12.3 When reading feature 1.7 in isolation, it may be linguistically correct to interpret the feature as specifying that a single figure of merit is commonly computed for all plausible combinations.

However, terms and features in a claim must be interpreted in context. The context to be considered for interpreting a feature in a claim not only includes the other features in that claim but also the other claims, the description and the drawings (see Case Law, II.A.6.3.2).

The board concurs with the appellant that computing a single figure of merit for the group of plausible combinations prevents the most probable combination from being determined since the information of the individual combinations is lost. Given that the aim is to determine an individual combination as a most probable combination, this interpretation does not make technical sense.

Moreover, paragraph [0038] of the patent specification supports the interpretation in which a figure of merit is computed for each plausible combination. It discloses computing a figure of merit for a triplet, for example, as an average of the matching errors for the triplet over the epipolar planes. A triplet is composed of the index of a curve in the first image, the index of a candidate corresponding curve in the second image and the index of a light sheet in the projector (see paragraph [0035]); that is, a plausible combination as specified in feature 1.4 of claim 1.

Paragraph [0038] also discloses that since the number of matching errors for a plausible combination over the epipolar planes depends on the visibility of the reflected curve in the two images, the curves of matching errors for plausible combinations shown in Figure 8 may have different lengths. Hence, the person skilled in the art does not interpret this passage to mean that a figure of merit is determined over only some of the epipolar planes, as submitted by respondent O2, but that it is determined over all epipolar planes, given that, due to the geometry of the objects to be scanned, not all plausible combinations in all epipolar planes would have a matching error.

3.12.4 The argument that Figure 9 supports a different interpretation of claim 1 is not convincing.

Figure 9 of the patent specification shows selecting an epipolar plane (920), identifying plausible triplet combinations along the selected epipolar plane (930) and calculating a figure of merit for each of the triplet combinations (940) within a loop. At the end of the loop, each image blob is associated with the most plausible triplet (950).

Claim 1 defines a loop over a set of epipolar planes including the steps of selecting an epipolar plane (feature 1.3), identifying plausible combinations (feature 1.4) and calculating a matching error for each of the plausible combinations (feature 1.5). Subsequently, a most probable combination is determined by computing a figure of merit for the plausible combinations (feature 1.7).

Hence, Figure 9 does not show calculating a matching error as defined by claim 1.

It appears to the board that the opposition division, in stating that "the disclosure in figure 9 falls within the scope of feature 1.7" (see point 10.2.5 of the decision under appeal), disregarded the context of feature 1.7 in claim 1, in particular the requirement that a matching error be calculated. Neither Figure 9 nor the related description in paragraph [0040] mentions the step of calculating a matching error. The board is not convinced by respondent O2's argument that calculating a matching error is implicitly disclosed because it is not apparent where in the context of Figure 9 this calculation would be performed. Thus, the cited passage and Figure 9 are not a proper basis for interpreting the features of claim 1.

3.12.5 The board is not persuaded by the argument that the same number of most probable combinations as light sheets should be identified in claim 1.

The wording of feature 1.7 does not specify determining more than one most probable combination. It is clear from the reference to the most probable combination in feature 1.8 that only one most probable combination is determined and nothing more is required to identify matching points from the most probable combination. The board notes that identifying a most probable combination for each light sheet would maximise the number of points in the images that could be matched. However, the claimed wording is not directed to this aspect.

3.12.6 Therefore, taking into account the context of the claim and the description, the person skilled in the art would adopt the interpretation identified in point 3.12 as the sole interpretation for feature 1.7.

3.13 The interpretation of feature 1.8

3.14 If a term used in a claim has a clear technical meaning, as a rule, the description cannot be used to interpret such a term in a different way (see Case Law, II.A.6.3.1).

3.15 Respondent O2 argued that, in view of the problem set out in paragraphs [0008] and [0011] and the aim of the invention to determine a three-dimensional surface, feature 1.8 could only be understood to mean identifying all the points on the light sheets of the projected pattern. This interpretation was supported by paragraph [0039], which disclosed determining points for the whole frame from the observations for each triplet.

3.16 The appellant submitted that paragraphs [0008] to [0011] of the patent specification disclosed the background art, that the claimed invention aimed at identifying matching points between two images and that feature 1.8 specified this matching. Once a most probable combination was identified, all points of the blobs in that combination were matched.

3.17 It is not apparent to the board why the clear formulation "identifying matching points ... from said most probable combination" should be interpreted in a completely different manner to include all points on the light sheets of the projected pattern, i.e. matching points on light sheets from other combinations. This is simply not claimed.

Feature 1.8 does not define matching points using a combination other than the most probable combination identified in feature 1.7. Nor does it consider light sheets or blob identifiers from other combinations to identify the matching points. Instead, feature 1.8 conveys the unambiguous teaching that matching points are calculated from the most probable combination. In these circumstances, the board is of the view that general statements in the description regarding problems in the state of the art cannot be used to alter the clear technical teaching of this feature.

4. Patent as granted - sufficiency of disclosure (Article 100(b) EPC)

4.1 The ground for opposition under Article 100(b) EPC prejudices the maintenance of the European patent as granted if the patent does not disclose the invention in a manner sufficiently clear and complete for it to be carried out by a person skilled in the art.

Under the case law of the boards of appeal, the skilled person may use their common general knowledge to supplement the information contained in the application. However, it must be possible to reproduce a claimed step using the original application documents without any inventive effort over and above the ordinary skills of a practitioner (see Case Law, sections II.C.4.1 and II.C.5.1).

4.2 To assess whether the invention in a European patent is sufficiently disclosed, the claimed subject-matter must first be determined by interpreting the claim (i.e. establishing the meaning of the claimed features) from the perspective of the person skilled in the art. The board carried out this exercise in section 3. above.

In a second step, it must be assessed whether the invention defined by that subject-matter is disclosed in a manner sufficiently clear and complete for it to be carried out by a person skilled in the art.

4.3 In the decision under appeal (see point 10.2.2), the opposition division concluded that feature 1.4 of claim 1 of the patent as granted provided a definition of plausible combinations. Hence, the opposition division found that feature 1.4 did not prejudice the maintenance of the patent under Article 100(b) EPC.

4.3.1 Respondent O2 submitted that, for disclosure, the whole application needed to be considered. Paragraphs [0035] and [0036] of the patent specification disclosed reducing the number of possible combinations on the basis of a matching error metric to arrive at "probable matches". However, claim 1 used this metric only after "plausible combinations" had already been determined. Hence, the patent did not disclose how to arrive at the plausible combinations of feature 1.4 independently of calculating the matching error (see section III.1 of respondent O2's reply of 21 March 2022).

4.3.2 The appellant concurred with the opposition division's findings. The plausible combinations were those whose blobs crossed the epipolar line, as defined in feature 1.4 (see section 2.2.1 of the statement of grounds of appeal). According to paragraphs [0035] and [0036] of the patent, the error could be used not only to limit the number of combinations to a number of plausible combinations but also to reduce the number of plausible combinations. Hence, these paragraphs did not disclose that the matching errors must be taken into account to determine the plausible combinations, and feature 1.4 of the invention was sufficiently disclosed. Moreover, the opponents' submissions supporting different meanings of the phrase "plausible combinations" amounted to an objection of lack of clarity rather than one of insufficiency of disclosure.

4.3.3 As set out in point 3.3 above with respect to feature 1.4, the phrase "plausible combinations" is to be interpreted as those combinations whose blobs cross the conjugate epipolar lines.

Using this interpretation, the patent specification, and in particular feature 1.4 of claim 1 as granted, discloses how to identify plausible combinations in a manner sufficiently clear and complete for this to be carried out by a person skilled in the art.

The board further notes that paragraphs [0035] and [0036] do not make it compulsory to take account of the matching errors for determining the plausible combinations.

4.4 Respondent O2 submitted that feature 1.6 provided for a different plausible combination to be identified for each epipolar plane and for a matching error to be calculated for each of the plausible combinations. Since matching errors were not calculated for all epipolar planes, it was impossible to associate plausible combinations across several epipolar planes. According to paragraph [0037] of the patent specification, the combinations were for all the planes (see section III.2 of its reply of 21 March 2022).

4.4.1 The board agrees with respondent O2 that different plausible combinations may be determined for each epipolar plane; only blobs that cross the conjugate epipolar line of the respective epipolar plane form plausible combinations for that epipolar plane (see feature 1.4). Thus, not all plausible combinations identified for an epipolar plane may have a matching error in all epipolar planes.

4.4.2 However, this would not be a bar to associating plausible combinations across multiple epipolar planes. The blobs are indexed with a label (see paragraph [0031] of the patent specification). The combinations include a respective unique identifier (index) for a blob from each of the two images and a light sheet label (see feature 1.4 and paragraph [0035] of the patent specification). For each epipolar plane, the subset of combinations for which blob indices cross the respective conjugate epipolar line are identified as plausible combinations. Thus, it is straightforward to associate corresponding (plausible) combinations across multiple epipolar planes by matching their two blob indices and the light sheet label. A figure of merit, such as an average matching error (see feature 1.7 and paragraph [0038] of the patent specification), can be computed over corresponding combinations without any difficulties.

4.5 With respect to feature 1.7, the opposition division noted that "the disclosure of the patent application should be enabling over the entire range of the claim". It argued that the disclosure in Figure 9 fell "within the scope of feature 1.7" of claim 1, finding that the contradicting disclosures of Figures 8 and 9 and corresponding paragraphs [0038] and [0040] of the patent specification "[led] to difficulties in discerning how the method would be implemented" (see point 10.2.5 of the decision under appeal).

4.5.1 The appellant submitted that paragraph [0038] disclosed how to compute a figure of merit and a most probable combination in a manner sufficiently clear and complete for this to be carried out by a person skilled in the art. It reiterated that claim 1 required only one most probable combination to be determined, asserting that the person skilled in the art would know how to identify most probable combinations for other light sheets using the disclosure of the application and common general knowledge.

4.5.2 According to respondent O2, if the claim could be interpreted in two different ways and the patent did not allow one interpretation to be excluded, then both interpretations had to meet the requirements of the EPC. This was explicitly stated in the Case Law concerning added matter (see II.E.1.3.9 c) and II.E.1.3.9 e)) and should apply to Article 83 EPC as well. Respondent O2 also referred to T 1404/05, submitting that if a claim was open to two interpretations, both had to be sufficiently disclosed. The mere fact that the description made it clear that one interpretation was the one intended did not mean that the claim could be treated as being confined to that interpretation.

Furthermore, when feature 1.7 was construed according to the interpretation set out in point 3.12, it was not clearly and sufficiently disclosed how to select a most probable combination from all plausible combinations using a figure of merit since neither paragraph [0038] nor Figure 8 of the patent specification showed this aspect (see the paragraph bridging pages 19 and 20 and the last paragraph on page 25 of respondent O2's reply of 21 March 2022). More specifically, paragraph [0008] stated that increasing the number of light sheets introduced ambiguities that could not be resolved. This was the case disclosed in paragraph [0038]. There had to be as many most probable combinations with a low figure of merit as there were light sheets. However, to resolve resulting ambiguities in that case, paragraph [0038] taught rejecting combinations. Consequently, the person skilled in the art did not know how to select a most probable combination.

4.5.3 Since the board finds that claim 1 allows for only one interpretation of feature 1.7 (see point 3.12 above), respondent O2's arguments based on two different interpretations are inconsequential.

The board concurs with respondent O2 that paragraph [0038] describes determining a figure of merit for each of the plausible combinations (see the paragraph bridging pages 21 and 22 of respondent O2's reply of 21 March 2022). This paragraph also identifies the triplet whose error is depicted at curve 806 as producing the best figure of merit - calculated as an average error for a given triplet - even if another curve reaches a lower local minimum error. The person skilled in the art would readily identify the goal of determining the combination (triplet) with the best figure of merit as finding the combination with the lowest average error. Hence, the patent discloses feature 1.7, and in particular how to determine a most probable combination, in a manner sufficiently clear and complete for it to be carried out by a person skilled in the art.

The invention defined in claim 1 does not specify determining the same number of most probable combinations as light sheets. Therefore, whether the patent as granted discloses how to determine the remaining combinations in a manner sufficiently clear and complete is not relevant for the assessment of the sufficiency of disclosure of the claimed invention.

4.6 With respect to feature 1.8, the opposition division found that the patent did not disclose how to determine a plurality of matching points when only one most probable combination was determined, as specified in feature 1.7 of claim 1 (see point 10.2.5 of the decision under appeal).

4.6.1 The board agrees with the appellant that only one combination is needed to identify a plurality of matching points. When construing feature 1.7 according to the interpretation set out in point 3.12 above, the step of determining a most probable combination provides a correspondence between a light sheet and respective blobs in each of the images, each blob being composed of a plurality of points (see paragraph [0031]), and the step of identifying matching points identifies the correspondence between coordinates of the points on the line of light of the light sheet and the points on the blobs (see the paragraph bridging pages 3 and 4 and the first full paragraph on page 4 of the statement of grounds of appeal).

4.6.2 In this context, respondent O2 contested that the patent disclosed that the matching points to be identified in feature 1.8 were points on the two blobs of the most probable combination and requested that the appellant's argument in this regard not be admitted into the proceedings for being late-filed.

The appellant submitted that the opposition division's reasoning in this respect was not clear from the discussion at the oral proceedings before the opposition division (see section 2.2.2 of the statement of grounds of appeal).

Indeed, it is not apparent from the minutes of the oral proceedings that the question of determining a plurality of matching points from a single most probable combination was discussed at all. Hence, the board admits the appellant's argument under Article 12(4) RPBA as it addresses a ground on which the decision was based and is not prejudicial to procedural economy. Moreover, the appellant's argument merely reflects the teaching of paragraph [0039] of the patent specification.

4.6.3 Therefore, the patent specification discloses the step of identifying matching points from the most probable combination in a manner sufficiently clear and complete for it to be carried out by a person skilled in the art.

4.7 In view of points 4.3 to 4.6 above, the board considers that the ground for opposition under Article 100(b) EPC does not prejudice the maintenance of the European patent as granted.

5. Remittal for further prosecution (Article 111 EPC)

5.1 The appellant requested that the board remit the case to the opposition division in view of the volume and complexity of the undecided issues and in view of Article 12(2) RPBA (see section 1.3 of the statement of grounds of appeal).

Respondents O1 and O2 argued that the undecided issues, in particular admittance of the requests, novelty and inventive step, should be dealt with before the board because all parties had presented their case on these issues. Remitting the case to the opposition division would be detrimental to procedural economy and would lead to a ping-pong effect between the board and the opposition division. The lengthy proceedings were a burden on the respondents.

Respondent O2 requested that the oral proceedings be adjourned so that the remaining issues on which the parties had presented their case could be discussed on a new date.

5.2 Under Article 111(1), second sentence, EPC the board may either exercise any power within the competence of the department which was responsible for the decision appealed or remit the case to that department for further prosecution.

Since the main purpose of the appeal proceedings is to give the losing party a chance to challenge the opposition division's decision on its merits (see G 10/91, OJ EPO 1993, 420, point 18 of the Reasons), the boards have normally considered remittal in accordance with Article 111(1) EPC to be appropriate in cases where the opposition division issues a decision on particular issues (e.g. added subject-matter and sufficiency of disclosure) but leaves other substantive issues (e.g. admissibility of claim requests and submissions, validity of the priority claim, novelty and inventive step) undecided. This existing practice fulfils the primary object of appeal proceedings to review the decision under appeal in a judicial manner, as expressed in Article 12(2) RPBA.

Although the EPC does not grant parties an absolute right to have all the issues in a case considered at two instances and a remittal would delay the proceedings, it is well recognised that any party may be given the opportunity to have two readings of the important elements of a case.

In the case in hand, the board disagrees with the opposition division's finding that claim 1 of the patent as granted could be interpreted in more than one way and finds that the invention as defined in claim 1 meets the requirements of Articles 83 and 123(2) EPC.

5.3 In the board's view, all these elements, and in particular the interpretation of claim 1 established by the board, constitute special reasons that justify remitting the case to the opposition division in accordance with Article 11 RPBA for an assessment of novelty and inventive step on the basis of that interpretation. Therefore, the board concludes that the case should be remitted to the opposition division for further prosecution. The request for adjournment of the oral proceedings then becomes moot.

6. Conclusion

6.1 Since the grounds for opposition under Article 100(b) and Article 100(c) EPC do not prejudice the maintenance of the European patent as granted, the decision under appeal is to be set aside. The case is to be remitted to the opposition division for further prosecution.