Research map / 3D vision / 3D generation
BlockGAN: Learning 3D Object-aware Scene Representations from Unlabelled Images
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| BlockGAN vs HoloGAN | 3D-aware GAN · CLEVR2 | KID: 0.032 -> 0.021 | Helped |
| BlockGAN vs LR-GAN | object-aware GAN · Real-Car | KID: 0.014 -> 0.016 | Hurt |
| element-wise max instead of sum composer (ablation) | BlockGAN · Synth-Chair1 | KID: 0.038 -> 0.031 | Helped |
| element-wise max instead of MLP composer (ablation) | BlockGAN · Synth-Car1 | KID: 0.044 -> 0.039 | Helped |
| use perspective camera instead of weak-perspective (ablation) | BlockGAN · synthetic datasets | qualitative disentanglement | Helped |
| add style discriminator (ablation) | BlockGAN · Real-Car | qualitative fg/bg separation | Helped |
| train with non-uniform true pose distribution under uniform pose prior (ablation) | BlockGAN · imbalanced Synth-Car1 | qualitative rotation control | Hurt |
Closest papers by meaning
- Learning 3D Scene Priors With 2D Supervision (CVPR 2023 · similarity 0.89)
- Learning 3D Object Shape and Layout Without 3D Supervision (CVPR 2022 · similarity 0.88)
- Scene Representation Networks: Continuous 3D-Structure-Aware Neural Scene Representations (NeurIPS 2019 · similarity 0.88)
- Neural Groundplans: Persistent Neural Scene Representations from a Single Image (ICLR 2023 · similarity 0.88)
- Disentangled 3D Scene Generation with Layout Learning (ICML 2024 · similarity 0.87)
- DORSal: Diffusion for Object-centric Representations of Scenes $\textit{et al.}$ (ICLR 2024 · similarity 0.87)
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.