AI Paper Atlas

Research map / Detection and segmentation / CNN features

TCFormer: Visual Recognition via Token Clustering Transformer

Wang Zeng, Sheng Jin, Lumin Xu, Wentao Liu, Chen Qian, Wanli Ouyang, Ping Luo, Xiaogang Wang

TPAMI 2024 · Detection and segmentation › CNN features · approach: Plain and high-resolution ViTs · 25 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
add global CTM token clustering in place of CNN merge (ablation)TCFormerV2-Small + Semantic FPN · ADE20K valmIoU: 45.2 -> 46.1Improved
add local CTM token clustering in place of CNN merge (ablation)TCFormerV2-Small + Semantic FPN · ADE20K valmIoU: 45.2 -> 45.7Improved
use local instead of global CTM (ablation)TCFormerV2-Small + Semantic FPN · ADE20K valGFLOPs of CTM: 42.1 GFLOPs -> 5.6 GFLOPsImproved
replace FPN with SR-MTA aggregation (local CTM) (ablation)TCFormerV2-Small + Semantic FPN · ADE20K valmIoU: 45.7 -> 47.1Improved
use CR-MTA instead of SR-MTA (local CTM) (ablation)TCFormerV2-Small + Semantic FPN · ADE20K valmIoU: 47.1 -> 47.8Improved
use CR-MTA instead of SR-MTA (global CTM) (ablation)TCFormerV2-Small + Semantic FPN · ADE20K valmIoU: 46.5 -> 47.0Improved
swap local DPC-kNN for K-means at test time without fine-tuning (ablation)TCFormerV2-Small · ImageNet-1K valtop-1 accuracy: 82.40 -> 82.39No clear effect
swap local DPC-kNN for Bipartite Soft Matching at test time (ablation)TCFormerV2-Small · ImageNet-1K valtop-1 accuracy: 82.40 -> 82.31No clear effect
use local instead of global clustering (TCFormerV2 vs V1)TCFormerV2-Small · ImageNet-1K valGFLOPs: 5.9 -> 4.5Improved
use dynamic-token TCFormerV2-Small instead of grid-token HRNet-w32COCO-WholeBody · COCO-WholeBody V1.0 valhand keypoint AP: 0.473 -> 0.535Improved
use TCFormerV2-Small instead of PVTv2-B2 for small objectsRetinaNet 1x · COCO val2017AP_S: 27.4 -> 28.8Improved
use TCFormerV2-Base instead of Swin-B with Mask2FormerMask2Former 640x640 · ADE20K valmIoU: 52.4 -> 53.8Improved

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