AI Paper Atlas

Research map / Vision-language and multimodal / Visual question answering

MoVie: Revisiting Modulated Convolutions for Visual Counting and Beyond

Duy Kien Nguyen, Vedanuj Goswami, Xinlei Chen

ICLR 2021 · Vision-language and multimodal › Visual question answering · approach: Attention and prior-robust VQA · 16 citations (OpenAlex)

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 changedWhereMeasuredEffect
Replace FiLM modulation with MoVie modulation (ablation)MoVie counting module · HowMany-QA valACC: 57.1 -> 58.4Improved
Add the learned linear mapping W to modulation (ablation)FiLM-style modulation · HowMany-QA valACC: 57.1 -> 58.4Improved
Add a residual connection to MoVie modulation (ablation)MoVie modulation · HowMany-QA valACC: 57.9 -> 58.4Improved
Add beta to MoVie modulation with the residual connection (ablation)MoVie modulation · HowMany-QA valACC: 58.4 -> 58.5No clear effect
Use multiple modulated bottlenecks instead of one (ablation)MoVie counting module · HowMany-QA valACC: 57.2 -> 58.4Improved
Use 5 rather than 4 modulated bottlenecks (ablation)MoVie counting module · HowMany-QA valRMSE: 2.60 -> 2.63Hurt
Pad images to a fixed maximum input size (ablation)MoVie counting module · HowMany-QA val, test size 800ACC: 56.2 -> 58.4Improved
Train at multiple image scales (ablation)MoVie counting module · HowMany-QA val, test size 800ACC: 58.4 -> 58.8Improved
Integrate MoVie using the three-branch fusion schemeMCAN-Small · VQA 2.0 valNumber accuracy: 46.71 -> 50.45Improved
Use naive feature fusion instead of three-branch fusion (ablation)MCAN-Small + MoVie · VQA 2.0 valNumber accuracy: 46.71 -> 49.36; 50.45 with three-branch fusionMixed
Add MoVie to MCAN-Large with ResNeXt-101MCAN-Large + X-101 · VQA 2.0 test-devNumber accuracy: 55.68 -> 57.05Improved
Add self-attention layers to question encoding (ablation)MoVie question encoder · HowMany-QA valACC: 0.7% improvement in accuracyImproved

Later papers that compared with this method

1 later accepted paper reports MoVie as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy