ITACLIP: Boosting Training-Free Semantic Segmentation with Image, Text, and Architectural Enhancements
This work addresses the need for better semantic segmentation in open-vocabulary computer vision tasks, but it is incremental as it builds on existing CLIP models with modifications.
The paper tackles the problem of improving dense prediction capabilities in Vision Language Models for Open-Vocabulary Semantic Segmentation by introducing architectural changes, image engineering, and text enhancements to CLIP, resulting in a training-free method that outperforms state-of-the-art approaches on benchmarks like COCO-Stuff and Pascal VOC.
Recent advances in foundational Vision Language Models (VLMs) have reshaped the evaluation paradigm in computer vision tasks. These foundational models, especially CLIP, have accelerated research in open-vocabulary computer vision tasks, including Open-Vocabulary Semantic Segmentation (OVSS). Although the initial results are promising, the dense prediction capabilities of VLMs still require further improvement. In this study, we enhance the semantic segmentation performance of CLIP by introducing new modules and modifications: 1) architectural changes in the last layer of ViT and the incorporation of attention maps from the middle layers with the last layer, 2) Image Engineering: applying data augmentations to enrich input image representations, and 3) using Large Language Models (LLMs) to generate definitions and synonyms for each class name to leverage CLIP's open-vocabulary capabilities. Our training-free method, ITACLIP, outperforms current state-of-the-art approaches on segmentation benchmarks such as COCO-Stuff, COCO-Object, Pascal Context, and Pascal VOC. Our code is available at https://github.com/m-arda-aydn/ITACLIP.