AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Models Matter, So Does Training: An Empirical Study of CNNs for Optical Flow Estimation

Deqing Sun, Xiaodong Yang, Ming-Yu Liu, Jan Kautz

TPAMI 2019 · Architectures and efficiency › RNNs and normalization · approach: Dataset condensation and flow estimation · 210 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 learned feature pyramid instead of image pyramid (ablation)PWC-Net · MPI Sintel Final train (trained on FlyingChairs)AEPE: 5.58 -> 4.59Improved
increase feature extractor to three layers per level (ablation)PWC-Net · MPI Sintel Final train (trained on FlyingChairs)AEPE: 4.59 -> 4.17Improved
add warping layer (ablation)PWC-Net · MPI Sintel Final train (trained on FlyingChairs)AEPE: 5.30 -> 4.59Improved
add context network (ablation)PWC-Net · MPI Sintel Final train (trained on FlyingChairs)AEPE: 4.74 -> 4.59Improved
add DenseNet connections (ablation)PWC-Net · MPI Sintel Final train (trained on FlyingChairs)AEPE: 4.37 -> 4.59Mixed
add cost volume (search range 4 vs 0) (ablation)PWC-Net · MPI Sintel Final train (trained on FlyingChairs)AEPE: 5.09 -> 4.59Improved
deepen optical flow estimator to seven layers (ablation)PWC-Net · MPI Sintel Final train (trained on FlyingChairs)AEPE: 4.59 -> 4.50Improved
make flow estimator residual (ablation)PWC-Net · MPI Sintel Final train (trained on FlyingChairs)AEPE: 4.59 -> 4.43Improved
use 6-level instead of 7-level pyramid (ablation)PWC-Net · MPI Sintel Final train (trained on FlyingChairs)AEPE: 4.59 -> 4.32Mixed
add FlyingThings3D fine-tuning after FlyingChairs (ablation)PWC-Net · MPI Sintel Final trainAEPE: 4.59 -> 3.93Improved
retrain FlowNetC with PWC-Net training procedure (FlowNetC+) vs published FlowNet2FlowNetC · MPI Sintel Final testAEPE: 6.02 -> 5.47Improved
fix KITTI I/O bug and retrain on all 200 sequencesPWC-Net-ft · KITTI 2015 testFl-all (%): 9.60% -> 7.90%Improved

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