AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Dynamic MLP for Fine-Grained Image Classification by Leveraging Geographical and Temporal Information

Lingfeng Yang, Xiang Li, Renjie Song, Borui Zhao, Juntian Tao, Shihao Zhou, Jiajun Liang, Jian Yang

CVPR 2022 · Vision-language and multimodal › Human and embodied modeling · approach: Scientific domain vision datasets · 51 citations (OpenAlex)

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
Use metadata-conditioned dynamic projection instead of image-only classificationDynamic MLP with SK-Res2Net-101 · iNaturalist 2021 Minitop-1 accuracy: 76.102 -> 84.694Improved
Use metadata-conditioned dynamic projection instead of the strongest listed multimodal baselineDynamic MLP with SK-Res2Net-101 · iNaturalist 2021 Fulltop-1 accuracy: 91.137 -> 91.397Improved
Use dynamic MLP instead of matched-capacity concatenation fusioniNaturalist 2021 Mini fusion comparison · iNaturalist 2021 MiniTop-1: 83.915 -> 84.694Improved
Use dynamic MLP instead of matched-capacity addition fusioniNaturalist 2021 Mini fusion comparison · iNaturalist 2021 MiniTop-1: 84.068 -> 84.694Improved
Use dynamic MLP instead of matched-capacity multiplication fusioniNaturalist 2021 Mini fusion comparison · iNaturalist 2021 MiniTop-1: 81.649 -> 84.694Improved
Add concatenated image and multimodal features before embedding in both paths (ablation)Dynamic MLP-C · iNaturalist 2021 Minitop-1 accuracy: 84.341 -> 84.694Improved
Add deeper embedding layers (ablation)Dynamic MLP · iNaturalist 2021 Minitop-1 accuracy: 84.341 -> 84.406Improved
Concatenate inputs only in the image path rather than both paths (ablation)Dynamic MLP-C · iNaturalist 2021 Minitop-1 accuracy: 84.694 -> 84.573Hurt
Concatenate inputs only in the multimodal path rather than both paths (ablation)Dynamic MLP-C · iNaturalist 2021 Minitop-1 accuracy: 84.694 -> 84.443Hurt
Use two dynamic projection stages rather than one (ablation)Dynamic MLP-A · iNaturalist 2021 Minitop-1 accuracy: 84.186 -> 84.341Improved
Use two dynamic projection stages rather than three (ablation)Dynamic MLP-A · iNaturalist 2021 Minitop-1 accuracy: 84.052 -> 84.341Improved
Use two dynamic projection stages rather than four (ablation)Dynamic MLP-A · iNaturalist 2021 Minitop-1 accuracy: 84.100 -> 84.341Improved

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