Research map / 3D vision / NeRF and radiance fields
FlowCam: Training Generalizable 3D Radiance Fields without Camera Poses via Pixel-Aligned Scene Flow
Cameron Omid Smith, Yilun Du, Ayush Tewari, Vincent Sitzmann
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 |
|---|---|---|---|
| use FlowCam instead of the strongest baseline | FlowCam · CO3D-Hydrants | PSNR: 18.81 -> 19.37 | Improved |
| use FlowCam instead of the strongest baseline | FlowCam · CO3D-10 | PSNR: 19.45 -> 21.14 | Improved |
| use FlowCam instead of the strongest baseline | FlowCam · RealEstate | PSNR: 17.83 -> 19.78 | Improved |
| use FlowCam instead of the strongest baseline | FlowCam · KITTI | PSNR: 15.17 -> 17.69 | Improved |
| use 3D-lifted flow pose estimation instead of VidAE | FlowCam · CO3D Hydrant | ATE: 1.8 -> 0.062 | Improved |
| use 3D-lifted flow pose estimation instead of VidAE | FlowCam · CO3D 10-Cat. | ATE: 0.66 -> 0.23 | Improved |
| use 3D-lifted flow pose estimation instead of VidAE | FlowCam · RE10K | ATE: 0.096 -> 0.012 | Improved |
| use 3D-lifted flow pose estimation instead of VidAE | FlowCam · KITTI | ATE: 0.12 -> 0.028 | Improved |
| replace direct CNN pose prediction with flow-based pose estimation (ablation) | FlowCam · CO3D Hydrant | PSNR: 18.50 -> 21.02 | Improved |
| replace direct CNN pose prediction with flow-based pose estimation (ablation) | FlowCam · CO3D Hydrant | LPIPS: 0.54 -> 0.38 | Improved |
| add flow confidence weights to Procrustes pose estimation (ablation) | FlowCam · CO3D Hydrant | PSNR: 17.87 -> 21.02 | Improved |
| add flow confidence weights to Procrustes pose estimation (ablation) | FlowCam · CO3D Hydrant | LPIPS: 0.51 -> 0.38 | Improved |
Later papers that compared with this method
3 later accepted papers report FlowCAM as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- SelfSplat: Pose-Free and 3D Prior-Free Generalizable 3D Gaussian Splatting (CVPR 2025 · better on 7 of 7 shared results)
- Unifying Correspondence Pose and NeRF for Generalized Pose-Free Novel View Synthesis (CVPR 2024 · better on 2 of 2 shared results)
- GGRt: Towards Generalizable 3D Gaussians without Pose Priors in Real-Time (ECCV 2024 · better on 1 of 1 shared result)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- RUST — RUST: Latent Neural Scene Representations From Unposed Imagery (CVPR 2023) · better on 8 of 8 shared results
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
- Flow-NeRF: Joint Learning of Geometry, Poses, and Dense Flow within Unified Neural Representations (CVPR 2025 · similarity 0.90)
- Joint Optimization of Neural Radiance Fields and Continuous Camera Motion from a Monocular Video (CVPR 2025 · similarity 0.88)
- 4D3R: Motion-Aware Neural Reconstruction and Rendering of Dynamic Scenes from Monocular Videos (NeurIPS 2025 · similarity 0.88)
- 3D Reconstruction with Generalizable Neural Fields using Scene Priors (ICLR 2024 · similarity 0.87)
- No Pose, No Problem: Surprisingly Simple 3D Gaussian Splats from Sparse Unposed Images (ICLR 2025 · similarity 0.87)
- Flux4D: Flow-based Unsupervised 4D Reconstruction (NeurIPS 2025 · 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.
