Summary of Sfc: Shared Feature Calibration in Weakly Supervised Semantic Segmentation, by Xinqiao Zhao et al.
SFC: Shared Feature Calibration in Weakly Supervised Semantic Segmentation
by Xinqiao Zhao, Feilong Tang, Xiaoyang Wang, Jimin Xiao
First submitted to arxiv on: 22 Jan 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
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 The proposed paper addresses the issue of long-tailed distribution in weakly supervised semantic segmentation, which can lead to degraded pseudo-label quality and final performance. Existing methods rely on Class Activation Mapping (CAM) for training semantic segmentation models, but this approach is limited by over-activation for head classes and under-activation for tail classes due to shared features. The authors introduce a Shared Feature Calibration (SFC) method that leverages class prototypes to calibrate the CAM generation process. This approach uses a Multi-Scaled Distribution-Weighted (MSDW) consistency loss to counterbalance over-activation and under-activation, leading to improved pseudo-label quality and state-of-the-art performance. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper is about finding a better way to do image segmentation without labeling every part of the image. This helps because it can be very time-consuming to label everything. Right now, people are using something called Class Activation Mapping (CAM) to get started, but this method has some problems. It can make mistakes for certain parts of the image and not others. The authors came up with a new idea called Shared Feature Calibration (SFC) that helps fix these problems. They tested it and found that it works really well. |
Keywords
» Artificial intelligence » Image segmentation » Semantic segmentation » Supervised
