Research map / Video / Video-language understanding
BANZ-FS: BANZSL Fingerspelling Dataset
Xin Shen, Yan Ke, Xinyu Wang, Xin Yu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use character-level tokenization (ByT5) instead of subword tokenization (T5) | Transformer translation model · BANZ-FS FSR-Context | Letter Accuracy (%): 10.2% -> 25.8% | Improved |
| train on the combined Full set instead of News only | HandReader · BANZ-FS IFSR (test Full) | Letter Accuracy (%): 60.8 -> 75.4 | Improved |
| train in-domain on Web instead of Lab data, tested on Web | HandReader · BANZ-FS IFSR (test Web) | Letter Accuracy (%): 32.6 -> 55.0 | Improved |
| train on Full instead of Lab only, tested on Lab | HandReader · BANZ-FS IFSR (test Lab) | Letter Accuracy (%): 93.1 -> 86.7 | Hurt |
| train on Full instead of Web only, tested on Web | FS-PoseNet · BANZ-FS IFSR (test Web) | Letter Accuracy (%): 51.4 -> 70.1 | Improved |
| use multimodal RGB + 3D pose input (HandReader) instead of pose-only FS-PoseNet | IFSR model · BANZ-FS IFSR (train Full, test Full) | Letter Accuracy (%): 74.7 -> 75.4 | Improved |
| use frame-level BIO tagging with pose cues (SL-Seg) instead of proposal-based Modified R-C3D | fingerspelling detector · BANZ-FS FSD (train Full, test Full) | AP@IoU0.5: 48.8 -> 66.9 | Improved |
| add recognition requirement (FSD-R) on top of detection | SL-Seg · BANZ-FS FSD to FSD-R | AP: 66.9 -> 53.5 | Hurt |
| train detector on Lab only instead of Web only, tested on Web | SL-Seg · BANZ-FS FSD (test Web) | AP@IoU0.5: 21.5 -> 40.4 | Hurt |
| use pose input instead of RGB input | SL-Luong · Auslan-Daily News V2 (single-person) | ROUGE: 15.47 -> 21.71 | Improved |
| crop the signer (single-person) instead of full frame (multi-person) | GFSLT-VLP · Auslan-Daily News V2 | ROUGE: 21.87 -> 26.24 | Improved |
| crop the signer (single-person) instead of full frame (multi-person) | GASLT · Auslan-Daily News V2 | ROUGE: 20.78 -> 23.94 | Improved |
Closest papers by meaning
- FSboard: Over 3 Million Characters of ASL Fingerspelling Collected via Smartphones (CVPR 2025 · similarity 0.92)
- MM-WLAuslan: Multi-View Multi-Modal Word-Level Australian Sign Language Recognition Dataset (NeurIPS 2024 · similarity 0.86)
- Lost in Translation, Found in Embeddings: Sign Language Translation and Alignment (NeurIPS 2026 · similarity 0.86)
- A Wireframe-Based Approach for Classifying and Acquiring Proficiency in the American Sign Language (Student Abstract) (AAAI 2024 · similarity 0.85)
- VSNet: Focusing on the Linguistic Characteristics of Sign Language (CVPR 2025 · similarity 0.85)
- LERMO: A Novel Web Game for AI-Enhanced Sign Language Recognition (AAAI 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
