Research map / Vision-language and multimodal / Visual question answering
Question-Guided Hybrid Convolution for Visual Question Answering
Peng Gao, Hongsheng Li, Shuang Li, Pan Lu, Yikang Li, Steven C.H. Hoi, Xiaogang Wang
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 |
|---|---|---|---|
| add group shuffling layer (ablation) | QGHC · VQA val | accuracy (All): 58.92 -> 59.24 | Improved |
| use 8 groups instead of 4 (ablation) | QGHC · VQA val | accuracy (All): 59.01 -> 59.24 | Improved |
| use 8 groups instead of 16 (ablation) | QGHC · VQA val | accuracy (All): 58.22 -> 59.24 | Improved |
| stack 3 modules instead of 1 (ablation) | QGHC · VQA val | accuracy (All): 58.88 -> 59.24 | Improved |
| stack 4 modules instead of 3 (ablation) | QGHC · VQA val | accuracy (All): 59.24 -> 59.13 | No clear effect |
| double output channels (vs halved) (ablation) | QGHC · VQA val | accuracy (All): 58.78 -> 59.24 | Improved |
| add residual 1x1-3x3-1x1 structure to naive full convolution (ablation) | QGHC-1-naive · VQA val | accuracy (All): 55.32 -> 57.01 | Improved |
| convert full convolutions to 8-group convolutions (ablation) | QGHC-1-full · VQA val | accuracy (All): 57.01 -> 58.41 | Improved |
| use hybrid (dynamic plus freely-updated) kernels instead of all-predicted group kernels (ablation) | QGHC-1-group · VQA val | accuracy (All): 58.41 -> 58.88 | Improved |
| concatenate QGHC features with question features (ablation) | QGHC · VQA val | accuracy (All): 59.24 -> 59.80 | Improved |
| fuse QGHC features with question features via MUTAN (ablation) | QGHC · VQA val | accuracy (All): 59.24 -> 60.13 | Improved |
| add attention (ablation) | QGHC · VQA val | accuracy (All): 59.24 -> 60.64 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- MUTAN — MUTAN: Multimodal Tucker Fusion for Visual Question Answering (ICCV 2017) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Multi-Level Attention Networks for Visual Question Answering (CVPR 2017 · similarity 0.92)
- Multi-Question Learning for Visual Question Answering (AAAI 2020 · similarity 0.92)
- Visual Question Answering as a Meta Learning Task (ECCV 2018 · similarity 0.91)
- An Analysis of Visual Question Answering Algorithms (ICCV 2017 · similarity 0.91)
- Tips and Tricks for Visual Question Answering: Learnings From the 2017 Challenge (CVPR 2018 · similarity 0.91)
- Question Type Guided Attention in Visual Question Answering (ECCV 2018 · similarity 0.91)
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.
