Research map / Detection and segmentation / CNN features
Embracing Events and Frames with Hierarchical Feature Refinement Network for Object Detection
Hu Cao, Zehua Zhang, Yan Xia, Xinyi Li, Jiahao Xia, Guang Chen, Alois C. Knoll
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 |
|---|---|---|---|
| fuse events with frames using CAFR instead of frames only (ablation) | RetinaNet · DSEC | mAP (%): 25.0 -> 38.0 | Improved |
| fuse events with frames using CAFR instead of frames only (ablation) | RetinaNet · PKU-DDD17-Car | mAP50 (%): 82.7 -> 86.7 | Improved |
| fuse events with frames using CAFR instead of frames only | RetinaNet · Corrupted PKU-DDD17-Car frames | mPC50 (%): 38.7 -> 69.5 | Improved |
| use CAFR instead of second-best EFNet fusion | RetinaNet · DSEC | mAP (%): 30.0 -> 38.0 | Improved |
| add CrossAtt and FR to the Mul&Add fusion in both branches (ablation) | CAFR · PKU-DDD17-Car | mAP (%): 41.5 -> 46.0 | Improved |
| add Mul&Add to CrossAtt plus FR (ablation) | CAFR · PKU-DDD17-Car | mAP (%): 45.3 -> 46.0 | Improved |
| use both frame- and event-dominated branches instead of a single branch (ablation) | CAFR · PKU-DDD17-Car | mAP (%): 44.7 -> 46.0 | Improved |
| combine Mul&Add with CrossAtt without FR (ablation) | CAFR · PKU-DDD17-Car | mAP50 (%): 84.3 -> 82.7 | Hurt |
| use event-only input instead of frame-only input (ablation) | RetinaNet · DSEC | mAP (%): 25.0 -> 12.0 | Hurt |
| use CAFR on corrupted blur frames versus RENet | RetinaNet · Corruption blur group | mPC50 (%): 72.3 (RENet) vs 57.0 (CAFR) | Hurt |
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.
- ECANet — ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks (CVPR 2020) · better on 1 of 1 shared result
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
- FlexEvent: Towards Flexible Event-Frame Object Detection at Varying Operational Frequencies (NeurIPS 2025 · similarity 0.88)
- Scene Adaptive Sparse Transformer for Event-based Object Detection (CVPR 2024 · similarity 0.87)
- Asymmetric Hierarchical Difference-aware Interaction Network for Event-guided Motion Deblurring (AAAI 2025 · similarity 0.86)
- Event-assisted Low-Light Video Object Segmentation (CVPR 2024 · similarity 0.85)
- Event Stream-based Visual Object Tracking: A High-Resolution Benchmark Dataset and A Novel Baseline (CVPR 2024 · similarity 0.85)
- Efficient Event-Based Semantic Segmentation via Exploiting Frame-Event Fusion: A Hybrid Neural Network Approach (AAAI 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.
