Research map / Video / Video segmentation and tracking
End-to-End Learning of Representations for Asynchronous Event-Based Data
Daniel Gehrig, Antonio Loquercio, Konstantinos G. Derpanis, Davide Scaramuzza
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 |
|---|---|---|---|
| learn the temporal kernel end-to-end instead of the trilinear kernel (ablation) | EST (time stamps) · N-Cars | accuracy: 0.917 -> 0.925 | Improved |
| learn the temporal kernel end-to-end instead of the trilinear kernel (ablation) | EST (time stamps) · N-Caltech101 | accuracy: 0.787 -> 0.817 | Improved |
| use handcrafted alpha kernel instead of trilinear kernel (ablation) | EST (time stamps) · N-Caltech101 | accuracy: 0.787 -> 0.739 | Hurt |
| use handcrafted exponential kernel instead of trilinear kernel (ablation) | EST (time stamps) · N-Cars | accuracy: 0.917 -> 0.909 | Hurt |
| learn the temporal kernel end-to-end instead of the trilinear kernel (ablation) | EST (time stamps) · MVSEC indoor_flying2 | AEE: 1.71 -> 1.38 | Improved |
| learn the temporal kernel end-to-end instead of the trilinear kernel (ablation) | EST (time stamps) · MVSEC indoor_flying3 | % outlier: 8.29 -> 6.47 | Improved |
| use alpha or exponential kernel instead of trilinear kernel (ablation) | EST (time stamps) · MVSEC indoor_flying2 | AEE: 1.71 -> 1.52 | Improved |
| retain temporal dimension (Voxel Grid) instead of summing over it (Two-Channel Image) (ablation) | event representation with time stamp measurements · MVSEC indoor_flying2 | AEE: 1.97 -> 1.70 | Improved |
| separate polarity instead of summing over polarities (ablation) | event representations · N-Cars / N-Caltech101 | accuracy: decrease in accuracy of up to 7% | Improved |
| use time stamps as measurement in EST instead of polarity (ablation) | EST (trilinear) · N-Caltech101 | accuracy: 0.789 -> 0.787 | No clear effect |
| increase temporal bins from 9 to 16 (ablation) | learned EST · N-Cars | accuracy: 0.925 -> 0.923 | No clear effect |
| increase temporal bins from 9 to 16 (ablation) | learned EST · N-Caltech101 | accuracy: 0.817 -> 0.837 | Improved |
Closest papers by meaning
- Event-based Asynchronous Sparse Convolutional Networks (ECCV 2020 · similarity 0.92)
- Events-To-Video: Bringing Modern Computer Vision to Event Cameras (CVPR 2019 · similarity 0.91)
- EventNet: Asynchronous Recursive Event Processing (CVPR 2019 · similarity 0.91)
- A Differentiable Recurrent Surface for Asynchronous Event-Based Data (ECCV 2020 · similarity 0.90)
- Graph Neural Network Combining Event Stream and Periodic Aggregation for Low-Latency Event-based Vision (CVPR 2025 · similarity 0.90)
- Leveraging Asynchronous Spiking Neural Networks for Ultra Efficient Event-Based Visual Processing (AAAI 2025 · similarity 0.90)
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.
