AI Paper Atlas

Research map / Architectures and efficiency / Neural architecture search

FlowNAS: Neural Architecture Search for Optical Flow Estimation

Zhiwei Lin, Tingting Liang, Taihong Xiao, Yongtao Wang, Ming–Hsuan Yang

IJCV 2023 · Architectures and efficiency › Neural architecture search · approach: Bi-level differentiable NAS · 8 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 SepConv search space instead of Conv (ablation)FlowNAS · KITTI-15 trainval (C+T)F1-all (%): 17.4 -> 16.9Improved
use SepConv search space instead of ShuffleConv (ablation)FlowNAS · KITTI-15 trainval (C+T)F1-all (%): 17.9 -> 16.9Improved
use Channel Maximize instead of Dynamic Channel Projection for feature alignment (ablation)FlowNAS FAD · KITTI-15 valF1-all (%): 4.80 -> 3.63Improved
use Channel Maximize instead of Spatial Attention (ablation)FlowNAS FAD · KITTI-15 valF1-all (%): 83.0 -> 3.63Improved
use Channel Maximize instead of Channel Average (ablation)FlowNAS FAD · KITTI-15 valF1-all (%): 3.89 -> 3.63Improved
add feature alignment distillation (FAD) to super-network training (ablation)FlowNAS-RAFT · KITTI-15 valF1-all (%): 4.05 -> 3.63Improved
add FAD when training sub-network from scratch (ablation)FlowNAS-RAFT · KITTI-15 valF1-all (%): 3.84 -> 3.78Improved
add rough decoder supervision L_rough (ablation)FlowNAS super-network · Sintel val FinalEPE: 3.00 -> 3.91Hurt
replace flow loss with rough decoder loss only (ablation)FlowNAS super-network · KITTI-15 valF1-all (%): 3.63 -> 98.7Hurt
add FAD loss L_D to flow loss (ablation)FlowNAS super-network · KITTI-15 valF1-all (%): 4.05 -> 3.63Improved
apply FlowNAS encoder search to GMAGMA · KITTI-15 trainval (C+T)F1-all (%): 17.1 -> 16.9Improved
apply FlowNAS-RAFT-S encoder to PWC-NetPWC-Net · Sintel trainval (C+T)EPE Clean: 2.55 -> 1.83Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 3 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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