Research map / 3D vision / 3D generation
GOEmbed: Gradient Origin Embeddings for Representation Agnostic 3D Feature Learning
Animesh Karnewar, Roman Shapovalov, Tom Monnier, Andrea Vedaldi, Niloy J. Mitra, David Novotny
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 changed | Where | Measured | Effect |
|---|---|---|---|
| train with diffusion with forward model driven by GOEmbed encodings instead of non-forward diffusion (ablation) | GOEmbedFusion · OmniObject3D generation | FID: 119.67 -> 22.12 | Improved |
| use GOEmbed-driven DFM instead of vanilla DFM | 3D generation · OmniObject3D generation | FID: 73.51 -> 22.12 | Improved |
| GOEmbed encoding instead of DinoV2 cost volume (triplane, 1 view) | Plenoptic Encoding · OmniObject3D Triplane | PSNR: 14.182 -> 15.711 | Improved |
| GOEmbed encoding instead of DinoV2 cost volume (voxel grid) | Plenoptic Encoding · OmniObject3D Voxel-Grid | PSNR: 15.834 -> 15.755 (4 views) | Mixed |
| increase number of source views from 1 to 4 in GOEmbed (ablation) | GOEmbed triplane · OmniObject3D Triplane | PSNR: 15.711 -> 15.919 | Improved |
| replace LRM with GOEmbed embeddings in reconstruction backbone | sparse-view reconstruction · OmniObject3D | PSNR: 23.788 -> 27.650 | Improved |
| use smaller 6-layer LRM instead of 16-layer (ablation) | LRM baseline · OmniObject3D | PSNR: 23.247 -> 23.788 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- DiffRF — DiffRF: Rendering-Guided 3D Radiance Field Diffusion (CVPR 2023) · better on 1 of 1 shared result
- GET3D — GET3D: A Generative Model of High Quality 3D Textured Shapes Learned from Images (NeurIPS 2022) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Multiview Compressive Coding for 3D Reconstruction (CVPR 2023 · similarity 0.87)
- Learning 3D Representations From 2D Pre-Trained Models via Image-to-Point Masked Autoencoders (CVPR 2023 · similarity 0.86)
- PonderV2: Improved 3D Representation With a Universal Pre-Training Paradigm (TPAMI 2025 · similarity 0.86)
- Object-X: Learning to Reconstruct Multi-Modal 3D Object Representations (NeurIPS 2025 · similarity 0.86)
- Sparse3D: Distilling Multiview-Consistent Diffusion for Object Reconstruction from Sparse Views (AAAI 2024 · similarity 0.85)
- ImageNet3D: Towards General-Purpose Object-Level 3D Understanding (NeurIPS 2024 · similarity 0.85)
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.
