Research map / Video / Video segmentation and tracking
Revisit Event Generation Model: Self-Supervised Learning of Event-to-Video Reconstruction with Implicit Neural Representations
Zipeng Wang, yunfan lu, Lin Wang
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 spatial regularization (ablation) | EvINR · IJRR | MSE: 0.063 -> 0.044 | Improved |
| add coarse-to-fine training (ablation) | EvINR · IJRR | training time (s): 17.29 -> 7.94 | Improved |
| add coarse-to-fine training (ablation) | EvINR · IJRR | MSE: 0.063 -> 0.061 | No clear effect |
| frame-based instead of coordinate-based optimization | EvINR · event sequences | convergence time: minutes -> seconds | Improved |
| ensemble 10 networks of 5 s each instead of one network (ablation) | EvINR · IJRR Calibration | MSE: 0.084 -> 0.0625 | Improved |
| increase ensemble partitions to 50 (1 s each) (ablation) | EvINR · IJRR Calibration | MSE: 0.0625 -> 0.0875 | Hurt |
| self-supervised EvINR versus supervised methods on a new sensor | EvINR · AED | SSIM: 0.345 -> 0.458 | Improved |
| EvINR versus best supervised method on lower-density events | EvINR · HQF | MSE: 0.032 vs 0.048 | Hurt |
| EvINR versus prior SSL method | EvINR · IJRR | MSE: 0.097 -> 0.047 | Improved |
Closest papers by meaning
- Back to Event Basics: Self-Supervised Learning of Image Reconstruction for Event Cameras via Photometric Constancy (CVPR 2021 · similarity 0.91)
- Sensing Diversity and Sparsity Models for Event Generation and Video Reconstruction from Events (TPAMI 2023 · similarity 0.89)
- EventSR: From Asynchronous Events to Image Reconstruction, Restoration, and Super-Resolution via End-to-End Adversarial Learning (CVPR 2020 · similarity 0.89)
- High Speed and High Dynamic Range Video with an Event Camera (TPAMI 2019 · similarity 0.88)
- E2HQV: High-Quality Video Generation from Event Camera via Theory-Inspired Model-Aided Deep Learning (AAAI 2024 · similarity 0.88)
- LaSe-E2V: Towards Language-guided Semantic-aware Event-to-Video Reconstruction (NeurIPS 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.
