AI Paper Atlas

Research map / Detection and segmentation / CNN features

P2T: Pyramid Pooling Transformer for Scene Understanding

Yu-Huan Wu, Yun Liu, Xin Zhan, Ming‐Ming Cheng

TPAMI 2022 · Detection and segmentation › CNN features · approach: Plain and high-resolution ViTs · 338 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
use four parallel pooling ratios instead of single ratio 8 (ablation)P2T-Small (ablation) · ImageNet-1Ktop-1 accuracy: 73.9 -> 74.7Improved
use four parallel pooling ratios instead of single ratio 8 (ablation)P2T-Small (ablation) · ADE20KmIoU: 34.4 -> 35.7Improved
use single large pooling ratio 24 instead of 8 (ablation)P2T-Small (ablation) · ImageNet-1Ktop-1 accuracy: 73.9 -> 70.6Hurt
apply multiple pooling at stage 4 on top of stages 2,3 (ablation)P2T-Small (ablation) · ImageNet-1Ktop-1 accuracy: 74.1 -> 74.5Improved
apply multiple pooling at all stages (ablation)P2T-Small (ablation) · ADE20KmIoU: 34.4 -> 35.7Improved
use average pooling instead of max pooling (ablation)P2T-Small (ablation) · ADE20KmIoU: 33.2 -> 35.7Improved
use average pooling instead of convolution (ablation)P2T-Small (ablation) · ImageNet-1Ktop-1 accuracy: 73.8 -> 74.7Improved
use fixed pooling ratios instead of fixed pooled sizes (ablation)P2T-Small (ablation) · ADE20KmIoU: 33.3 -> 35.7Improved
use fixed pooled sizes (saves compute) (ablation)P2T-Small (ablation) · ImageNet-1KGFlops: 41.6 -> 38.9Mixed
add relative positional encoding (ablation)P2T-Small (ablation) · ADE20KmIoU: 35.7 -> 37.4Improved
add IRB after RPE (ablation)P2T-Small (ablation) · ImageNet-1Ktop-1 accuracy: 76.4 -> 79.5Improved
add overlapping patch embedding after RPE and IRB (ablation)P2T-Small (ablation) · ADE20KmIoU: 42.7 -> 44.1Improved

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