A reinforcement learning agent that controls SAM 2 memory bank updates achieves a +4.91% tracking quality gain over SAM 2 when overfitted per video, indicating untapped potential in memory control.
Segment A nything
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
SAM2RL: Towards Reinforcement Learning Memory Control in Segment Anything Model 2
A reinforcement learning agent that controls SAM 2 memory bank updates achieves a +4.91% tracking quality gain over SAM 2 when overfitted per video, indicating untapped potential in memory control.