AI Paper Atlas

Research map / Video / Video segmentation and tracking

From Shadow Segmentation to Shadow Removal

Hieu Le, Dimitris Samaras

ECCV 2020 · Video › Video segmentation and tracking · approach: Mask propagation for unsupervised video segmentation · 116 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 bounded search space for shadow parameters w and b (ablation)Ours · ISTD adjusted testRMSE LAB shadow area: 47.5 -> 9.7Improved
add boundary loss L_bd (ablation)Ours · ISTD adjusted testRMSE LAB shadow area: 41.7 -> 9.7Improved
add matting loss L_mat-alpha (ablation)Ours · ISTD adjusted testRMSE LAB shadow area: 38.7 -> 9.7Improved
add smoothness loss L_sm-alpha (ablation)Ours · ISTD adjusted testRMSE LAB shadow area: 10.2 -> 9.7Improved
add adversarial loss L_GAN (ablation)Ours · ISTD adjusted testRMSE LAB shadow area: 26.9 -> 9.7Improved
add boundary loss L_bd (ablation)Ours · ISTD adjusted testRMSE LAB all: 9.8 -> 4.0Improved
add matting loss L_mat-alpha (ablation)Ours · ISTD adjusted testRMSE LAB all: 9.0 -> 4.0Improved
train on unpaired shadow patches with only shadow masks instead of unpaired shadow-free images (MaskShadow-GAN)Ours · ISTD adjusted testRMSE LAB shadow area: 12.4 -> 9.7Improved
use paired shadow-free supervision (SP+M-Net) instead of mask-only trainingSP+M-Net · ISTD adjusted testRMSE LAB shadow area: 7.9 -> 9.7Mixed
fine-tune on each test video for 1 epoch with detector-generated masksOurs · Video Shadow Removal datasetRMSE LAB on moving-shadow mask: 20.9 -> 18.0Improved
use patch-based mask-only model rather than SP+M-Net on videosSP+M-Net · Video Shadow Removal datasetRMSE LAB on moving-shadow mask: 22.2 -> 20.9Improved

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