SAM2RL: Towards Reinforcement Learning Memory Control in Segment Anything Model 2
By: Alen Adamyan , Tomáš Čížek , Matej Straka and more
Potential Business Impact:
Teaches computers to follow moving things better.
Segment Anything Model 2 (SAM 2) has demonstrated strong performance in object segmentation tasks and has become the state-of-the-art for visual object tracking. The model stores information from previous frames in a memory bank, enabling temporal consistency across video sequences. Recent methods augment SAM 2 with hand-crafted update rules to better handle distractors, occlusions, and object motion. We propose a fundamentally different approach using reinforcement learning for optimizing memory updates in SAM 2 by framing memory control as a sequential decision-making problem. In an overfitting setup with a separate agent per video, our method achieves a relative improvement over SAM 2 that exceeds by more than three times the gains of existing heuristics. These results reveal the untapped potential of the memory bank and highlight reinforcement learning as a powerful alternative to hand-crafted update rules for memory control in visual object tracking.
Similar Papers
Rethinking Memory Design in SAM-Based Visual Object Tracking
CV and Pattern Recognition
Helps computers remember objects better when they disappear.
MoSAM: Motion-Guided Segment Anything Model with Spatial-Temporal Memory Selection
CV and Pattern Recognition
Helps computers track moving things in videos.
Memory-Augmented SAM2 for Training-Free Surgical Video Segmentation
CV and Pattern Recognition
Helps robots see and track tools in surgery.