SAM-CLIP: Merging Vision Foundation Models towards Semantic and Spatial Understanding

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Wang, Haoxiang, Vasu, Pavan Kumar Anasosalu, Faghri, Fartash, Vemulapalli, Raviteja, Farajtabar, Mehrdad, Mehta, Sachin, Rastegari, Mohammad, Tuzel, Oncel, Pouransari, Hadi
Format: Preprint
Publié: 2023
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866911912221999104
author Wang, Haoxiang
Vasu, Pavan Kumar Anasosalu
Faghri, Fartash
Vemulapalli, Raviteja
Farajtabar, Mehrdad
Mehta, Sachin
Rastegari, Mohammad
Tuzel, Oncel
Pouransari, Hadi
author_facet Wang, Haoxiang
Vasu, Pavan Kumar Anasosalu
Faghri, Fartash
Vemulapalli, Raviteja
Farajtabar, Mehrdad
Mehta, Sachin
Rastegari, Mohammad
Tuzel, Oncel
Pouransari, Hadi
contents The landscape of publicly available vision foundation models (VFMs), such as CLIP and Segment Anything Model (SAM), is expanding rapidly. VFMs are endowed with distinct capabilities stemming from their pre-training objectives. For instance, CLIP excels in semantic understanding, while SAM specializes in spatial understanding for segmentation. In this work, we introduce a simple recipe to efficiently merge VFMs into a unified model that absorbs their expertise. Our method integrates techniques of multi-task learning, continual learning, and distillation. Further, it demands significantly less computational cost compared to traditional multi-task training from scratch, and it only needs a small fraction of the pre-training datasets that were initially used to train individual models. By applying our method to SAM and CLIP, we obtain SAM-CLIP: a unified model that combines the capabilities of SAM and CLIP into a single vision transformer. Compared with deploying SAM and CLIP independently, our merged model, SAM-CLIP, reduces storage and compute costs for inference, making it well-suited for edge device applications. We show that SAM-CLIP not only retains the foundational strengths of SAM and CLIP, but also introduces synergistic functionalities, notably in zero-shot semantic segmentation, where SAM-CLIP establishes new state-of-the-art results on 5 benchmarks. It outperforms previous models that are specifically designed for this task by a large margin, including +6.8% and +5.9% mean IoU improvement on Pascal-VOC and COCO-Stuff datasets, respectively.
format Preprint
id arxiv_https___arxiv_org_abs_2310_15308
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle SAM-CLIP: Merging Vision Foundation Models towards Semantic and Spatial Understanding
Wang, Haoxiang
Vasu, Pavan Kumar Anasosalu
Faghri, Fartash
Vemulapalli, Raviteja
Farajtabar, Mehrdad
Mehta, Sachin
Rastegari, Mohammad
Tuzel, Oncel
Pouransari, Hadi
Computer Vision and Pattern Recognition
Machine Learning
The landscape of publicly available vision foundation models (VFMs), such as CLIP and Segment Anything Model (SAM), is expanding rapidly. VFMs are endowed with distinct capabilities stemming from their pre-training objectives. For instance, CLIP excels in semantic understanding, while SAM specializes in spatial understanding for segmentation. In this work, we introduce a simple recipe to efficiently merge VFMs into a unified model that absorbs their expertise. Our method integrates techniques of multi-task learning, continual learning, and distillation. Further, it demands significantly less computational cost compared to traditional multi-task training from scratch, and it only needs a small fraction of the pre-training datasets that were initially used to train individual models. By applying our method to SAM and CLIP, we obtain SAM-CLIP: a unified model that combines the capabilities of SAM and CLIP into a single vision transformer. Compared with deploying SAM and CLIP independently, our merged model, SAM-CLIP, reduces storage and compute costs for inference, making it well-suited for edge device applications. We show that SAM-CLIP not only retains the foundational strengths of SAM and CLIP, but also introduces synergistic functionalities, notably in zero-shot semantic segmentation, where SAM-CLIP establishes new state-of-the-art results on 5 benchmarks. It outperforms previous models that are specifically designed for this task by a large margin, including +6.8% and +5.9% mean IoU improvement on Pascal-VOC and COCO-Stuff datasets, respectively.
title SAM-CLIP: Merging Vision Foundation Models towards Semantic and Spatial Understanding
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2310.15308