Research map / Video / Video segmentation and tracking
Bayesian Tracking of Video Graphs Using Joint Kalman Smoothing and Registration
Aditi Basu Bal, Ramy Mounir, Sathyanarayanan Aakur, Sudeep Sarkar, Anuj Srivastava
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 |
|---|---|---|---|
| add backward smoothing pass to Kalman filter (ablation) | graph Kalman filter · Synthetic Dataset 1 | node L2 error: 0.009 -> 0.008 | Improved |
| add backward smoothing pass to Kalman filter (ablation) | graph Kalman filter · Synthetic Dataset 1 | edge L2 error: 0.091 -> 0.091 | No clear effect |
| add backward smoothing pass to Kalman filter (ablation) | graph Kalman filter · Synthetic Dataset 2 | node L2 error: 0.008 -> 0.006 | Improved |
| add backward smoothing pass to Kalman filter (ablation) | graph Kalman filter · Synthetic Dataset 2 | edge L2 error: 0.084 -> 0.085 | No clear effect |
| replace median filter with Kalman smoother | median filter · Synthetic Dataset 1 | node L2 error: 0.146 -> 0.008 | Improved |
| replace GRU multi-epoch training with Kalman smoother | GRU · Synthetic Dataset 2 | edge L2 error: 0.493 -> 0.085 | Improved |
| add backward smoothing pass to Kalman filter (ablation) | graph Kalman filter · MEVA clip 1 | rate of false detections: 0.721 -> 0.762 | Hurt |
| add backward smoothing pass to Kalman filter (ablation) | graph Kalman filter · MEVA clip 2 | rate of false detections: 0.997 -> 0.956 | Improved |
| add backward smoothing pass to Kalman filter (ablation) | graph Kalman filter · MEVA clip 3 | rate of missed detections: 3.580 -> 3.556 | Improved |
| replace median filter with Kalman smoother | median filter · MEVA clip 1 | rate of false detections: 4.118 -> 0.762 | Improved |
Closest papers by meaning
- TD²-Net: Toward Denoising and Debiasing for Video Scene Graph Generation (AAAI 2024 · similarity 0.85)
- Motion Graph Unleashed: A Novel Approach to Video Prediction (NeurIPS 2024 · similarity 0.85)
- Multi-Scene Generalized Trajectory Global Graph Solver with Composite Nodes for Multiple Object Tracking (AAAI 2024 · similarity 0.85)
- Towards Unbiased and Robust Spatio-Temporal Scene Graph Generation and Anticipation (CVPR 2025 · similarity 0.84)
- Sampling-Resilient Multi-Object Tracking (AAAI 2024 · similarity 0.84)
- TrackCraft3R: Repurposing Video Diffusion Transformers for Dense 3D Tracking (NeurIPS 2026 · similarity 0.84)
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.
