AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Learning Non-Lambertian Object Intrinsics Across ShapeNet Categories

Jian Shi, Yue Dong, Hao Su, Stella X. Yu

CVPR 2017 · Detection and segmentation › Open-set and few-shot detection · approach: Amodal and open-world instance segmentation · 192 citations (OpenAlex)

Check 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 the mirror-link encoder-decoder CNN with separate decoders for albedo, shading, and specularNon-Lambertian object intrinsics · ShapeNet intrinsics test setalbedo DSSIM: 0.1649 -> 0.0939Improved
Train Direct Intrinsics on ShapeNet-IntrinsicsDirect Intrinsics · ShapeNet intrinsics test setalbedo DSSIM: 0.1984 -> 0.1655Improved
Fine-tune the proposed model on MIT dataOurs · MIT intrinsics datasetalbedo DSSIM: 0.1825 -> 0.1465Improved
Fine-tune the proposed model on MIT dataOurs · MIT intrinsics datasetshading DSSIM: 0.1667 -> 0.1200Improved
Train on a category-specific dataset instead of all categoriesNon-Lambertian object intrinsics CNN · ShapeNet categoriesDSSIMImproved
Freeze the encoder and fine-tune only the decoder on another category (ablation)Category-trained intrinsics CNN · Cross-category ShapeNet testDSSIMNo clear effect
Use encoder features learned on one category to predict intrinsics in other categoriesNon-Lambertian object intrinsics CNN · Cross-category ShapeNet testDSSIMNo clear effect

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