Summary of Embracing Events and Frames with Hierarchical Feature Refinement Network For Object Detection, by Hu Cao et al.
Embracing Events and Frames with Hierarchical Feature Refinement Network for Object Detection
by Hu Cao, Zehua Zhang, Yan Xia, Xinyi Li, Jiahao Xia, Guang Chen, Alois Knoll
First submitted to arxiv on: 17 Jul 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Robotics (cs.RO)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes a novel hierarchical feature refinement network for event-frame fusion, addressing the issue of limited sensing capability in conventional cameras. The proposed method, called cross-modality adaptive feature refinement (CAFR), consists of two parts: bidirectional cross-modality interaction (BCI) and two-fold adaptive feature refinement (TAFR). The BCI part facilitates information bridging from two distinct sources, while the TAFR part refines features by aligning channel-level mean and variance. Experimental results on the PKU-DDD17-Car and DSEC datasets show that the proposed method surpasses state-of-the-art performance by 8.0% on the DSEC dataset and exhibits better robustness to corruption types. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper solves a problem in computer vision where cameras have trouble detecting objects in certain conditions. It proposes a new way to combine information from two different camera types: event cameras that output sparse events, and frame cameras that capture images. The new method is called the cross-modality adaptive feature refinement (CAFR) network. It works by first bringing together information from both camera types, then refining it to make sure it’s accurate. The results show that this new method does a better job than previous methods at detecting objects and can handle noisy or corrupted images. |
