analiza aiqbench
This research proposes a novel open-vocabulary framework for Japanese Sign Language recognition that bypasses traditional closed-set classification by generating procedural descriptions and retrieving them via sentence encoders. By fine-tuning vision-language models, the method significantly improves retrieval accuracy for both seen and unseen signs without requiring gloss supervision, effectively addressing generalization limits inherent in standard classifiers. This approach represents a meaningful advancement in machine learning applications for assistive technology, moving sign language processing toward scalable, unsupervised deployment.
A Reverse Sign Language Dictionary: Open-Vocabulary Sign Recognition from Continuous Signing via Video Captioning and Description Retrieval
arXiv:2609.03788v1 Announce Type: cross Abstract: Isolated Sign Language Recognition (ISLR) is conventionally cast as closed-set classification over gloss labels, which cannot generalize to signs unseen in training and ties every deployment to a gloss-annotated lexicon. We instead recognize signs extracted from continuous signing by (1) captioning a sign-level clip into a free-form procedural description of the articulation with an open-weight vision-language model, and (2) retrieving the closest entry from a vocabulary of target descriptions with a multilingual sentence encoder: a reverse sign language dictionary that needs no gloss supervision and admits an open vocabulary. On 1,300 sign-level segments from a Japanese Sign Language (JSL) dialogue corpus annotated with procedural descriptions (against a 2% top-10 chance floor over the 503-entry target vocabulary), fine-tuning the captioner substantially improves seen-class retrieval: language and vision tower fine-tuning raises top-10 retrieval on seen classes from 4.5% (untrained) to 49%, becoming statistically indistinguishable from a standard supervised closed-set classifier (I3D) on two of the three test sets where a closed-set classifier can be evaluated at all. More importantly, unseen-class retrieval also improves significantly over the untrained pipeline (11.5% -> 21.0% top-10, p=0.0094), a regime in which the closed-set classifier cannot participate. A matcher-side empirical upper-bound analysis shows the sentence encoder already recovers close to 100% of paraphrased gold descriptions, locating a gap in captioning quality that we aim to address in future work. To our knowledge this is the first description-based, open-vocabulary sign lookup from continuous signing without gloss supervision, and the first for JSL.
Abstract:Isolated Sign Language Recognition (ISLR) is conventionally cast as closed-set classification over gloss labels, which cannot generalize to signs unseen in training and ties every deployment to a gloss-annotated lexicon. We instead recognize signs extracted from continuous signing by (1) captioning a sign-level clip into a free-form procedural description of the articulation with an open-weight vision-language model, and (2) retrieving the closest entry from a vocabulary of target descriptions with a multilingual sentence encoder: a reverse sign language dictionary that needs no gloss supervision and admits an open vocabulary. On 1,300 sign-level segments from a Japanese Sign Language (JSL) dialogue corpus annotated with procedural descriptions (against a 2% top-10 chance floor over the 503-entry target vocabulary), fine-tuning the captioner substantially improves seen-class retrieval: language and vision tower fine-tuning raises top-10 retrieval on seen classes from 4.5% (untrained) to 49%, becoming statistically indistinguishable from a standard supervised closed-set classifier (I3D) on two of the three test sets where a closed-set classifier can be evaluated at all. More importantly, unseen-class retrieval also improves significantly over the untrained pipeline (11.5% -> 21.0% top-10, p=0.0094), a regime in which the closed-set classifier cannot participate. A matcher-side empirical upper-bound analysis shows the sentence encoder already recovers close to 100% of paraphrased gold descriptions, locating a gap in captioning quality that we aim to address in future work. To our knowledge this is the first description-based, open-vocabulary sign lookup from continuous signing without gloss supervision, and the first for JSL.
4 pages, 2 figures, 1 table. Extended version of an abstract presented at the BU-SHI workshop (Broadening the Users: A Cross-Disciplinary Roadmap for Social Humanoid Interaction), IEEE RO-MAN 2026, Kitakyushu, Japan, 28 August 2026. The workshop is non-archival; no proceedings
Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL)
arXiv:2609.03788 [cs.CV]
(or arXiv:2609.03788v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2609.03788 arXiv-issued DOI via DataCite (pending registration)
Text integral sub licența deschisă a publicației · 309 cuvinte
Subiecte conexe
Six Principles of Voice Design: Designing for the Ear and Encoding It into Agent Script
Salesforce Blog · 8 oct. 2026
Isolated Sign Language Recognition for Icelandic Sign Language: Experiments in a Low-resource Setting
arXiv · 23 sept. 2026
Token-Level Entropy Reveals Demographic Disparities in Large Language Models
arXiv · 25 iul. 2026
AWED-PIPER: Agents, Web Applications & Expert Detectors for Personally Identifiable Information Protection & Fine-grained Named Entity Recognition across 36 languages for 6.6 Billion Speakers
arXiv · 18 aug. 2026