AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Unsupervised Learning of Optical Flow with Deep Feature Similarity

Woobin Im, Tae-Kyun Kim, Sung-Eui Yoon

ECCV 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 34 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 occlusion handling to RGB photometric loss (ablation)PWC-Net with feature separation loss · FlyingChairs testEPE: 4.01 -> 3.64Improved
use census transform instead of RGB features (ablation)PWC-Net with feature separation loss · FlyingChairs testEPE: 3.64 -> 3.10Improved
add data distillation Ld to census loss (ablation)PWC-Net with feature separation loss · FlyingChairs testEPE: 3.10 -> 2.93Improved
add feature separation loss Lf and regulated census loss Lr on top of Ld (ablation)PWC-Net with feature separation loss · FlyingChairs testEPE: 2.93 -> 2.69Improved
add feature separation loss Lf and regulated census loss Lr on top of Ld (ablation)PWC-Net with feature separation loss · Sintel Final ALL (trained on Sintel)EPE: 3.86 -> 3.57Improved
add Lf without Lr (ablation)PWC-Net with feature separation loss · SintelEPENo clear effect
add regulated census loss Lr to the loss (ablation)PWC-Net with feature separation loss · Sintel / FlyingChairsEPEImproved
learn deep features with triplet loss instead of proposed losses (ablation)PWC-Net with feature separation loss · Sintel / FlyingChairsEPENo clear effect
use large smoothness weight 10^-2 instead of 10^-4 (ablation)PWC-Net with feature separation loss · FlyingChairsEPE: 3.27 vs. 2.85Hurt
raise separation weight lambda_f to 5.0 from 4.0 (ablation)PWC-Net with feature separation loss · FlyingChairsEPE: 3.21 vs. 2.85Hurt
add Lf and Lr for large displacements 40+ px (ablation)PWC-Net with feature separation loss · Sintel Final (trained on Sintel)EPE: 27.17 -> 22.36Improved
fine-tune on KITTI with proposed losses vs DDFlowPWC-Net with feature separation loss · KITTI 2015 testFl: 14.29% -> 13.38%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