AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Transformer-Based Attention Networks for Continuous Pixel-Wise Prediction

Guanglei Yang, Hao Tang, Mingli Ding, Nicu Sebe, Elisa Ricci

ICCV 2021 · Architectures and efficiency › Attention and transformers · approach: Efficient and robust vision transformers · 238 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 attention gate decoder (AGD) (ablation)ResNet-50 baseline · KITTI (Eigen split)delta<1.25: 0.888 -> 0.953Improved
add ViT Transformer layers (ablation)ResNet-50 baseline · KITTI (Eigen split)delta<1.25: 0.888 -> 0.955Improved
add both AGD and ViT (ablation)ResNet-50 baseline · KITTI (Eigen split)delta<1.25: 0.888 -> 0.956Improved
add AGD (ablation)ResNet-50 baseline · NYU Depth v2delta<1.25: 0.866 -> 0.881Improved
add ViT Transformer layers (ablation)ResNet-50 baseline · NYU Depth v2delta<1.25: 0.866 -> 0.887Improved
add both AGD and ViT (ablation)ResNet-50 baseline · NYU Depth v2delta<1.25: 0.866 -> 0.900Improved
add emitting feature f4 to f5 fusion (ablation)AGD · NYU Depth v2 (fusion ablation)delta<1.25: 0.878 -> 0.897Improved
add emitting feature f3 (three scales) (ablation)AGD · NYU Depth v2 (fusion ablation)delta<1.25: 0.897 -> 0.900Improved
add fourth emitting feature f2 (ablation)AGD · NYU Depth v2 (fusion ablation)rel: 0.106 -> 0.107Hurt
self-attention gating with only f5 as emitting and receiving feature (ablation)AGD · NYU Depth v2 (fusion ablation)delta<1.25: 0.866 -> 0.878Improved
use hybrid R50+ViT-B/16 instead of pure ViT-B/16 (ablation)TransDepth encoder · NYU Depth v2 (backbone ablation)delta<1.25: 0.885 -> 0.900Improved
use ViT-B/16 instead of ViT-B/32 on ResNet-50 (ablation)hybrid encoder · NYU Depth v2 (backbone ablation)delta<1.25: 0.893 -> 0.900Improved

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