--- title: "Scalable Video-Based Search in the — Vandendriessche 2026" description: "Searching a sign language dictionary by recording a sign has been studied in research but had not been deployed for a large vocabulary." authors: ["Toon Vandendriessche", "Caro Brosens", "Hannes De Durpel", "Mathieu De Coster", "Joni Dambre"] year: 2026 venue: "Proceedings of the 26th Annual Conference of the European Association for Machine Translation (Volume 2)" source: "https://aclanthology.org/2026.eamt-2.15/" licence: "CC BY-ND 4.0 (EAMT)" licence_url: "https://creativecommons.org/licenses/by-nd/4.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/en/papers/vandendriessche-2026-scalable-video-based-search-vgt/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/vandendriessche-2026-scalable-video-based-search-vgt/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/8/) / Scalable Video-Based Search in the VGT Dictionary # Scalable Video-Based Search in the VGT Dictionary **Authors:** Toon Vandendriessche, Caro Brosens, Hannes De Durpel, Mathieu De Coster, Joni Dambre · **Year:** 2026 · **Venue:** Proceedings of the 26th Annual Conference of the European Association for Machine Translation (Volume 2) · **Licence:** [CC BY-ND 4.0 (EAMT)](https://creativecommons.org/licenses/by-nd/4.0/) **Links:** [Publisher's page](https://aclanthology.org/2026.eamt-2.15/) · [PDF](https://aclanthology.org/2026.eamt-2.15.pdf) · [OpenAlex](https://openalex.org/W7165512813) **Citations:** 0 citations (openalex-snapshot · 2026-10-14) ## Summary Searching a sign language dictionary by recording a sign has been studied in research but had not been deployed for a large vocabulary. This paper reports such a deployment: a video-based lookup feature added to the Flemish Sign Language (VGT) Dictionary, which holds 11,248 unique signs and continues to grow. Before this, users could only search by sign parameters such as handshape or body location. The system was developed jointly by the deaf-led VGTC and Ghent University researchers. It retrieves signs through similarity search over generic vector representations, so new dictionary entries need no retraining. For privacy, keypoints are extracted on the user's device and no personal data leaves it. The recording pipeline was rebuilt to process every frame, and checks were added for whether the user's shoulders and elbows are in frame. Both changes improved retrieval on weaker hardware. The tool went live on 28 November 2025 and has seen steady weekly use. It makes the dictionary usable in both directions, which especially helps deaf and hard-of-hearing users with limited Dutch. The authors urge researchers to regard putting systems into real-world use as a worthwhile subject of study, not merely an afterthought. ## Key points - The search covers the full VGT Dictionary of 11,248 signs and has been publicly available since 28 November 2025. The authors describe it as the first video-based sign search at this vocabulary scale. - Retrieval uses similarity search over vector representations not specific to VGT, so the dictionary can grow without retraining. - MediaPipe extracts keypoints locally on the user's device for privacy. Two changes improved accuracy on low-end hardware: - forcing every frame to be processed, instead of real-time extraction that dropped frames; - detecting whether shoulders and elbows are in frame. - Optional user feedback after each query is meant to support long-term evaluation. So far only 172 labelled queries exist, which is too few for statistically solid conclusions. - The project was an equal partnership between the deaf-led VGTC and Ghent University (imec, IDLab AIRO). A companion paper by Vandendriessche et al. (2026) describes the co-creation process. *Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.* Fields: [Linguistics & languages](https://md.research.elamur.ai/free/en/shelves/linguistics/) · [Machine translation](https://md.research.elamur.ai/free/en/shelves/machine-translation/) ## Cite The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @inproceedings{vandendriessche2026scalable, title = {{Scalable Video-Based Search in the VGT Dictionary}}, author = {Vandendriessche, Toon and Brosens, Caro and Durpel, Hannes De and Coster, Mathieu De and Dambre, Joni}, booktitle = {Proceedings of the 26th Annual Conference of the European Association for Machine Translation (Volume 2)}, year = {2026}, url = {https://aclanthology.org/2026.eamt-2.15/} } ``` - [← previous SAVI: Web-based Multilayered Semantic Annotation Validation Interface](https://md.research.elamur.ai/free/en/papers/tatavolu-2026-savi-web-based-multilayered-semantic/) - [next → Scare Quotes as Markers of “Questionable” Word Usages and Misalignment in Conversation: An Annotation Study](https://md.research.elamur.ai/free/en/papers/soler-2026-scare-quotes-as-markers-questionable/) - [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/8/) ## References 5 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. Bragg, D. et al. 2019. Sign language recognition, generation, and translation: An interdisciplinary perspective. In Proceedings of the 21st international ACM SIGACCESS conference on computers and accessibility, pages 16–31. [find](https://scholar.google.com/scholar?q=Bragg%2C+D%2E+et+al%2E+2019%2E+Sign+language+recognition%2C+generation%2C+and+translation%3A+An+interdisciplinary+perspective%2E+In+Proceedings+of+the+21st+international+ACM+SIGACCESS+conference+on+computers+and+acce) 2. Brosens, C. et al. 2022. Moving towards a functional approach in the flemish sign language dictionary making process. In Proceedings of the LREC2022 10th Workshop on the Representation and Processing of Sign Languages: Multilingual Sign Language Resources, pages 24–28. [find](https://scholar.google.com/scholar?q=Brosens%2C+C%2E+et+al%2E+2022%2E+Moving+towards+a+functional+approach+in+the+flemish+sign+language+dictionary+making+process%2E+In+Proceedings+of+the+LREC2022+10th+Workshop+on+the+Representation+and+Processing+) 3. Sutton, V. 2022. Lessons in SignWriting. SignWriting Press. [find](https://scholar.google.com/scholar?q=Sutton%2C+V%2E+2022%2E+Lessons+in+SignWriting%2E+SignWriting+Press%2E) 4. Vandendriessche, T. et al. 2026. Signbuddy: from sign language research to scalable co-created solutions. Universal Access in the Information Society, 25(2):45. [find](https://scholar.google.com/scholar?q=Vandendriessche%2C+T%2E+et+al%2E+2026%2E+Signbuddy%3A+from+sign+language+research+to+scalable+co%2Dcreated+solutions%2E+Universal+Access+in+the+Information+Society%2C+25%282%29%3A45%2E) 5. Vermeerbergen, M. and M. Van Herreweghe. 2018. Looking back while moving forward: The impact of societal and technological developments on flemish sign language lexicographic practices. International Journal of Lexicography, 31(2):167–195. [find](https://scholar.google.com/scholar?q=Vermeerbergen%2C+M%2E+and+M%2E+Van+Herreweghe%2E+2018%2E+Looking+back+while+moving+forward%3A+The+impact+of+societal+and+technological+developments+on+flemish+sign+language+lexicographic+practices%2E+International+)