AlignSAM: Aligning Segment Anything Model to Open Context via Reinforcement Learning.
Duojun HuangXinyu XiongJie MaJichang LiZequn JieLin MaGuanbin LiPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- probabilistic model
- high level
- least squares
- computational model
- context dependent
- theoretical analysis
- mathematical model
- spatial context
- data sets
- conceptual framework
- formal model
- experimental data
- statistical model
- theoretical framework
- parameter estimation
- contextual information
- cost function
- multi agent
- objective function
- knowledge base
- information retrieval
- machine learning