Segment and Caption Anything
Preprint
Abstract
We propose a method to efficiently equip the Segment Anything Model (SAM) with the ability to generate regional captions. SAM presents strong generalizability to segment anything while is short for semantic understanding. By introducing a lightweight query-based feature mixer, we align the region-specific features with the embedding space of language models for later caption generation. As the number of trainable parameters is small (typically in the order of tens of millions), it costs less computation, less memory usage, and less communication bandwidth, resulting in both fast and scalable training. To address the scarcity problem of regional caption data, we propose to first pre-train our model on objection detection and segmentation tasks. We call this step weak supervision pretraining since the pre-training data only contains category names instead of full-sentence descriptions. The weak supervision pretraining allows us to leverage many publicly available object detection and segmentation datasets. We conduct extensive experiments to demonstrate the superiority of our method and validate each design choice. This work serves as a stepping stone towards scaling up regional captioning data and sheds light on exploring efficient ways to augment SAM with regional semantics. The project page, along with the associated code, can be accessed via https://xk-huang.github.io/segment-caption-anything/.Keywords
All Related Versions
This publication has 52 references indexed in Scilit:
- GPT4RoI: Instruction Tuning Large Language Model on Region-of-Interest2025
- GRiT: A Generative Region-to-Text Transformer for Object Understanding2024
- Exploring Plain Vision Transformer Backbones for Object Detection2022
- Detecting Twenty-Thousand Classes Using Image-Level Supervision2022
- DaViT: Dual Attention Vision Transformers2022
- Scaling Open-Vocabulary Image Segmentation with Image-Level Labels2022
- Learning Quality-aware Dynamic Memory for Video Object Segmentation2022
- Global Spectral Filter Memory Network for Video Object Segmentation2022
- End-to-End Object Detection with Transformers2020
- Microsoft COCO: Common Objects in Context2014