AI Paper Atlas

Research map / Video / Video segmentation and tracking

The Making and Breaking of Camouflage

Hala Lamdouar, Weidi Xie, Andrew Zisserman

ICCV 2023 · Video › Video segmentation and tracking · approach: Video restoration with temporal priors · 26 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 intra-image Frechet loss to the camouflage GAN generator (ablation)StyleGAN camouflage generator · synthetic camouflage imagescamouflage score S_alpha: 0.546 -> 0.598Improved
add intra-image Frechet loss to the GAN (ablation)StyleGAN camouflage generator · synthetic camouflage imagesFID: 75.01 -> 76.34Hurt
add intra-image Frechet loss to the GAN (ablation)StyleGAN camouflage generator · synthetic camouflage imagesInception Score (object clarity): 3.49 -> 3.13Hurt
pre-train on synthetic data then fine-tune on MoCA-Mask instead of MoCA-Mask only (ablation)full motion segmentation model · MoCA-Mask testmIoU: 23.4 -> 30.8Improved
add synthetic data to MoCA-Mask training of the simple model (ablation)baseline model without appearance encoder · MoCA-Mask testmIoU: 15.3 -> 16.0Improved
add appearance encoder (ablation)motion segmentation model · MoCA-Mask testmIoU: 16.0 -> 20.5Improved
add transformer encoder on top of the appearance encoder (ablation)motion segmentation model · MoCA-Mask testmIoU: 20.5 -> 23.0Improved
add transformer aggregation to the model with transformer encoder (ablation)motion segmentation model · MoCA-Mask testmIoU: 23.0 -> 30.8Improved
train on synthetic data only (ablation)baseline model · MoCA-Mask testmIoU: 14.5 vs 15.3 (MM only)Hurt
combine S_Rf and S_b (alpha=0.35) instead of either alone (ablation)camouflage score · CHAMELEON human rankingKendall-tau: 0.01 / 0.03 -> 0.42Improved
use boundary score instead of reconstruction fidelity (ablation)camouflage score · Camouflaged cuboids human rankingKendall-tau: -0.07 -> 0.42Improved
extend synthetic images to video sequencessynthetic camouflage generation · synthetic camouflagecamouflage score S_alpha: 0.598 -> 0.578Hurt

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