Research map / 3D vision / Depth and stereo
InFlux++: Real and Synthetic Data for Estimating Dynamic Camera Intrinsics
Erich Liang, Caleb Kha-Uong, Chinmaya Saran, Sreemanti Dey, David Liu, Junhan Ouyang, Benjamin Zhou, Jia Deng
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 |
|---|---|---|---|
| finetune AnyCalib on InFlux++ Synth | AnyCalib · InFlux+InFlux++ Real test (combined) | fx recall@10%: 25.2 -> 31.2 | Improved |
| finetune AnyCalib on InFlux++ Synth | AnyCalib · InFlux+InFlux++ Real test (combined) | fx recall@20%: 41.3 -> 50.4 | Improved |
| finetune AnyCalib on InFlux++ Synth | AnyCalib · InFlux test | fx recall@10%: 11.5 -> 17.5 | Improved |
| finetune AnyCalib on InFlux++ Synth | AnyCalib · InFlux++ Real test | fx recall@10%: 28.2 -> 34.2 | Improved |
| finetune AnyCalib on InFlux++ Synth | AnyCalib · InFlux+InFlux++ Real test (combined) | cx recall@0.5%: 41.6 -> 33.0 | Hurt |
| finetune AnyCalib on InFlux++ Synth | AnyCalib · InFlux+InFlux++ Real test (combined) | cy recall@1%: 37.4 -> 33.4 | Hurt |
| finetune AnyCalib on InFlux++ Synth | AnyCalib · InFlux+InFlux++ Real test (combined) | EPE recall@50px: 25.1 -> 18.7 | Hurt |
| finetune AnyCalib on InFlux++ Synth | AnyCalib · InFlux+InFlux++ Real test (combined) | EPE recall@300px: 87.3 -> 78.6 | Hurt |
| finetune AnyCalib on InFlux++ Synth | AnyCalib · InFlux++ Real test | EPE recall@10px: 6.40e-3 -> 1.66e-1 | Improved |
| finetune AnyCalib on InFlux++ Synth | AnyCalib · InFlux test | EPE recall@300px: 80.1 -> 82.7 | Improved |
| evaluate AnyCalib on InFlux++ Real instead of InFlux | AnyCalib · fx recall@10% | fx recall@10%: 11.5 -> 28.2 | Improved |
| evaluate AnyCalib on InFlux++ Real instead of InFlux | AnyCalib · cx recall@0.5% | cx recall@0.5%: 73.5 -> 34.7 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.
- AnyCalib — AnyCalib: On-Manifold Learning for Model-Agnostic Single-View Camera Calibration (ICCV 2025) · better on 5 of 14 shared results
- GeoCalib — GeoCalib: Learning Single-image Calibration with Geometric Optimization (ECCV 2024) · better on 2 of 2 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
- InFlux: A Benchmark for Self-Calibration of Dynamic Intrinsics of Video Cameras (NeurIPS 2025 · similarity 0.95)
- Depth Any Video with Scalable Synthetic Data (ICLR 2025 · similarity 0.85)
- DGS-LRM: Real-Time Deformable 3D Gaussian Reconstruction From Monocular Videos (NeurIPS 2025 · similarity 0.85)
- 4D3R: Motion-Aware Neural Reconstruction and Rendering of Dynamic Scenes from Monocular Videos (NeurIPS 2025 · similarity 0.85)
- Rethinking the State Update Gate for Long-Sequence Recurrent 3D Reconstruction (NeurIPS 2026 · similarity 0.85)
- Reconstruct, Inpaint, Test-Time Finetune: Dynamic Novel-view Synthesis from Monocular Videos (NeurIPS 2025 · 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.
