2 papers
cs.CV2024
Self-Supervised Vision Transformers Are Efficient Segmentation Learners for Imperfect Labels
Seungho Lee, Seoungyoon Kang, Hyunjung Shim
This study demonstrates a cost-effective approach to semantic segmentation using self-supervised vision transformers (SSVT). By freezing the SSVT backbone and training a lightweigh…
cs.CV2024
Weakly Supervised Semantic Segmentation for Driving Scenes
Dongseob Kim, Seungho Lee, Junsuk Choe +1
State-of-the-art techniques in weakly-supervised semantic segmentation (WSSS) using image-level labels exhibit severe performance degradation on driving scene datasets such as City…