AI Paper Atlas

Research map / 3D vision / 3D generation

Leveraging Multi-View Image Sets for Unsupervised Intrinsic Image Decomposition and Highlight Separation

Renjiao Yi, Ping Tan, Stephen Lin

AAAI 2020 · 3D vision › 3D generation · approach: Diffusion-based novel view synthesis · 37 citations (OpenAlex)

This paper tried a common change: Add a contrastive loss

add contrastive loss for joint finetuning · ShapeNet and real · qualitative Improved

What 1,317 accepted papers report for this change

See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.

Read the paper

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 unsupervised finetuning on real multi-view photos (ablation)Highlight separation network · Real lab datasetMSE (diffuse): 0.0045 -> 0.0036Improved
add unsupervised finetuning on real photos (ablation)Highlight separation network · ShapeNet Intrinsic syntheticMSE (diffuse): 0.0015 -> 0.0016Hurt
use color distribution loss instead of pixel-to-pixel low-rank loss (ablation)Highlight separation network · Real lab datasetMSE (diffuse): 0.0041 -> 0.0036Improved
use color distribution loss instead of pixel-to-pixel low-rank loss (ablation)Highlight separation network · ShapeNet Intrinsic syntheticMSE (diffuse): 0.0020 -> 0.0016Improved
add unsupervised finetuning (ablation)Intrinsic decomposition network · ShapeNet IntrinsicMSE (albedo): 0.0108 -> 0.0054Improved
use color distribution loss instead of pixel-to-pixel low-rank loss (ablation)Intrinsic decomposition network · ShapeNet IntrinsicMSE (shading): 0.0087 -> 0.0045Improved
use color distribution loss instead of pixel-to-pixel low-rank loss (ablation)Intrinsic decomposition network · ShapeNet IntrinsicMSE (albedo): 0.0067 -> 0.0054Improved
add contrastive loss for joint finetuning (ablation)Full H-Net + S-Net · ShapeNet and realqualitativeImproved
add supervised pretraining on 1.1% ShapeNet synthetic data (ablation)Fully unsupervised network · Real lab datasetMSE (diffuse): 0.0057 -> 0.0036Improved
add pretraining on synthetic data (ablation)Fully unsupervised network · ShapeNet IntrinsicMSE (albedo): 0.0067 -> 0.0054Improved
skip connections between encoder and decoder in H-Net (ablation)H-Net · ShapeNetqualitativeNo 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