AI Paper Atlas

Research map / Detection and segmentation / CNN features

TDAM: Top-Down Attention Module for Contextually Guided Feature Selection in CNNs

Shantanu Jaiswal, Basura Fernando, Cheston Tan

ECCV 2022 · Detection and segmentation › CNN features · approach: Convolution-transformer backbones · 17 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 TDjoint (t=2, m=1) top-down attention moduleResNet50 · ImageNet-V1 valtop-1 accuracy: 77.51 -> 78.96Improved
add TDjoint (t=2, m=1) top-down attention moduleResNet50 · ImageNet-V2 valtop-1 accuracy: 66.39 -> 67.66Improved
use TDjoint instead of FCA-TS attentionResNet50 · ImageNet-V1 valtop-1 accuracy: 78.70 -> 78.96Improved
use TDjoint (t=2, m=1) instead of SEResNet101 · ImageNet-V1 valtop-1 accuracy: 80.84 -> 81.62Improved
apply TD module only at layer 4 plus alternating blocks (L4) instead of layers 3 and 4 (ablation)ResNet101 TDjoint · ImageNet-V1 valtop-1 accuracy: 81.62 -> 81.12Hurt
increase feedback distance from m=1 to m=3 in TDtop (ablation)ResNet50 · ImageNet-V1 valtop-1 accuracy: 78.82 -> 78.90No clear effect
add top-down attention module for test resolution 448x448ResNet50 · ImageNet-V1 val (448x448 test)top-1 accuracy: 2% better than original ResNetImproved
use feedback distance of at least 1 (distinct top and bottom maps) (ablation)TD module on ResNet50 · ImageNet-V1 valtop-1 accuracyImproved
use channel-then-spatial attention instead of parallel channel and spatial (ablation)TD module on ResNet50 · ImageNet-V1 valtop-1 accuracyImproved
add spatial attention to channel-only feedback (ablation)TD module on ResNet50 · ImageNet-V1 valtop-1 accuracyImproved
use top-down attention instead of recurrent feedback convolution (ablation)TD module on ResNet50 · ImageNet-V1 valtop-1 accuracyImproved
increase feedback steps beyond t=2 (ablation)TD module · ImageNet-200 subsettop-1 accuracyHurt

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