T 1632/22 - Starting from some undefined method

Key points

• The applicant, in the first communication after the start of examination, amends the preamble of claim 1 from the original "a method of determining whether a target in a target image is a true image" by adding "thus detecting liveness of the target". The method, incidentally, is done using a neural network. The added feature is based on page 6, lines 11-12 of the description.
• "Example embodiments provide technology of accurately detecting a target using a different convolutional neural network based on a quality type of a target image regardless of various factors that affect the quality type of the target image. Here, detecting the target may be understood to determine whether the target is a true target, for example, to determine whether a target in the target image is a biometric target. For example, if the target included in the target image is a biometric face, the target may be recognized as a true target. If the target included in the target image is a non-biometric face, for example, a photograph and an image, the target may be recognized as a false target. Hereinafter, the target detection method may be understood as a method of detecting liveness of a target."
• The Examining Division refused the application for lack of inventive step over D1.
• "in the Board's judgement, the person skilled in the art in liveness detection would have regard to D1 and would have reason to adapt its solution to liveness detection in a way leading to the invention according to claim 1 of the main request before the Examining Division."
• "The current request differs from that request by defining the quality parameter as a combination of photographic and attribute parameters."
• The amendment was made in reply to some clarity objections raised by the Board. 
• "D1 discusses "attribute" parameters in the sense of the claim, [but] it does not discuss "photographic" parameters."
• The Board remits the case: " The Board notes that during examination a relatively large number of documents were cited, some of them concerned with liveness detection, but were not discussed in the decision. A positive decision on inventive step cannot be issued before at least these documents have been discussed."
• The Board, however, agrees with the Examining Division that claim 1 without the additional features is obvious over D1.
• "Document D1 is concerned with face authentication. It explains that when users try to authenticate themselves in conditions different to the ones where the registration took place, the accuracy may suffer (paragraph 8). To address this problem, D1 proposes the use of different classifiers for different condi­tions. Conditions include face orientation, accessories worn by user (e.g. glasses), but also illumination conditions and "image quality" (paragraphs 20 and 21). A condition-deriving unit detects the current condi­tions (paragraph 71) and a corresponding classifier is selected for authentication. The classifiers may be, inter alia, neural networks (paragraph 79)."
• The difference is "its use, i.e. (face) liveness detection instead of face authentication". Hence, the claim under examination defines the same concept of having different neural networks used for different image properties, but for "liveness detection" instead of "face authentication"
• "In the present case, the Board assumes a person skilled in liveness detection methods. The Board considers that such a person is, generally, interested in improving, or finding alternatives to, known liveness detection methods, based on the knowledge that known methods have known pitfalls"
• "Liveness detection for authentication and authentica­tion are closely related technical areas; the Appellant did not contest this. Therefore, a person skilled in liveness detection will naturally also have regard to prior art from the field of authentication for develop­ments relevant to liveness detection. This includes D1."
• "This skilled person will recognize that the conditions mentioned in D1, e.g. face orientation, accessories worn by user, and in particular illumination conditions (D1, paragraphs 20 and 21) have an influence on liveness detection methods as well."
• As a comment, the Board seems to assume some existing liveness method as the starting point for the inventive step assessment. The Board does not explain the precise nature of the existing liveness detection method to be developed. Apparently, the existing method is affected by light illumination, and this was known. Given that the Board also notes that the Examining Division had cited prior art about liveness detection, a specific reference would have been useful to just make clear (and certain) what existing method is the starting point. From my own experience: sometimes the prior art that should exist just does not want to be found during a search.
•  "The skilled person will also recognize that the solution proposed in D1, i.e classifiers specialized for different conditions can be applied in a straightforward manner for liveness detection, simply by, as the Examining Division stated, "do[ing] the same" for liveness detection."
• Again, without having a well-defined pre-existing liveness detection method, this is easy but risky to assume.
•   The Board notes in passing that it is a rather typical method of research to try adapting developments in neighbouring fields to the own area of interest. It is certainly common practice in image processing, in particular when the images are of the same type."
• As a comment, it is also not unusual for research to result in patentable inventions.
• "in the Board's judgement, the person skilled in the art in liveness detection would have regard to D1 and would have reason to adapt its solution to liveness detection in a way leading to the invention according to claim 1 of the main request before the Examining Division."
• Note again that the Board assumes existing "liveness detection" without telling us the details or citing a specific prior art document disclosing that liveness detection method.
• The priority date is in 2016. I suppose the awareness of deepfakes among the general public postdates the paradigm shift of the ChatGPT demo (30 November 2022). 
• The term "liveness detection" gives plenty of results in Google, but all after 2022 (except in the context of fingerprints). In fact, Ngram gives zero results for the usage up to 2022 in the Google Books corpus. 
• The Board includes a headnote and some general reasoning on the PSA. However, as of yet, it seems a one-off decision to me. 

EPO 

The link to the decision can be found after the jump.

https://www.epo.org/en/boards-of-appeal/decisions/t221632eu1