Research map / Vision-language and multimodal / Visual question answering
Grounding Answers for Visual Questions Asked by Visually Impaired People
Chongyan Chen, Samreen Anjum, Danna Gurari
Check 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 |
|---|---|---|---|
| Train MAC-Caps on VizWiz-VQA rather than GQA, CLEVR, or VQA-v2 | MAC-Caps · VizWiz-VQA-Grounding test | Avg IoU: 12.56 -> 27.43; 15.31 -> 27.43; 17.42 -> 27.43 | Improved |
| Train LXMERT on VizWiz-VQA rather than OSCAR on VQA-v2 | VQA grounding models · VizWiz-VQA-Grounding test | Avg IoU: 15.48 -> 22.09 | Improved |
| Use MAC-Caps trained on VizWiz-VQA rather than LXMERT trained on VizWiz-VQA | VizWiz-pretrained grounding models · VizWiz-VQA-Grounding test | Avg IoU: 22.09 -> 27.43 | Improved |
| Predict the whole image as the grounded area instead of using a learned grounding model | VizWiz-VQA-Grounding · VizWiz-VQA-Grounding test | Avg IoU: 27.43 -> 33% | Improved |
| Evaluate only questions whose answers the model predicted correctly | MAC-Caps pretrained on VizWiz-VQA · VizWiz-VQA-Grounding test | Avg IoU: 27.43 -> 32.8 | Improved |
| Ground small answer regions rather than medium or large ones | LXMERT, OSCAR, and MAC-Caps pretrained on VizWiz · VizWiz-VQA-Grounding test, small image coverage | IoU | Hurt |
| Evaluate on images with no quality issues rather than images with medium or severe issues | LXMERT and OSCAR · VizWiz-VQA-Grounding test, image quality | IoU: 0.28, 0.47 and 0.83 | Hurt |
| Ground questions requiring text recognition or counting rather than object or color recognition | LXMERT, OSCAR, and MAC-Caps pretrained on VizWiz · VizWiz-VQA-Grounding test, vision skills | IoU | Hurt |
| Pretrain grounding models on VizWiz-VQA rather than contrived VQA datasets | VQA and answer-grounding models · VizWiz-VQA-Grounding test | mAP: 0.01% -> 0.96%; 0.02% -> 0.96%; 0.03% -> 0.96%; 0.08% -> 0.96% | Improved |
Closest papers by meaning
- VQA Therapy: Exploring Answer Differences by Visually Grounding Answers (ICCV 2023 · similarity 0.94)
- VQA: Visual Question Answering (IJCV 2016 · similarity 0.93)
- VizWiz Grand Challenge: Answering Visual Questions From Blind People (CVPR 2018 · similarity 0.92)
- When Open-Vocabulary Visual Question Answering Meets Causal Adapter: Benchmark and Approach (AAAI 2025 · similarity 0.89)
- Can I Trust Your Answer? Visually Grounded Video Question Answering (CVPR 2024 · similarity 0.89)
- Right this way: Can VLMs Guide Us to See More to Answer Questions? (NeurIPS 2024 · similarity 0.89)
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.
