Summary of Semantic Prompt Learning For Weakly-supervised Semantic Segmentation, by Ci-siang Lin et al.
Semantic Prompt Learning for Weakly-Supervised Semantic Segmentation
by Ci-Siang Lin, Chien-Yi Wang, Yu-Chiang Frank Wang, Min-Hung Chen
First submitted to arxiv on: 22 Jan 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Computation and Language (cs.CL); Machine Learning (cs.LG)
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 SemPLeS framework for weakly-supervised semantic segmentation (WSSS) aims to train segmentation models using image-level supervision. The existing methods focus on producing pseudo masks by refining CAM-like heatmaps, but these may not accurately capture the object categories or backgrounds. To address this issue, SemPLeS learns to prompt the CLIP latent space for enhanced semantic alignment between segmented regions and target object categories. It uses Contrastive Prompt Learning and Prompt-guided Semantic Refinement to learn prompts that describe and suppress co-occurring backgrounds associated with each object category. This framework achieves competitive performance on standard WSSS benchmarks, PASCAL VOC 2012 and MS COCO 2014, and shows compatibility with other WSSS methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The SemPLeS framework helps computers better understand images by grouping similar objects together. It does this without needing detailed labels for every pixel in the image. Instead, it uses a clever trick to figure out what’s going on in the picture. This approach is useful because getting detailed labels can be very time-consuming and difficult. The SemPLeS framework works well on standard image datasets and can even work with other methods that try to do the same thing. |
Keywords
* Artificial intelligence * Alignment * Latent space * Prompt * Semantic segmentation * Supervised
