MoDiPO: text-to-motion alignment via AI-feedback-driven Direct Preference Optimization.
Massimiliano PappaLuca ColloroneGiovanni FicarraIndro SpinelliFabio GalassoPublished in: CoRR (2024)
Keyphrases
- artificial intelligence
- motion analysis
- camera motion
- information retrieval
- expert systems
- motion estimation
- image sequences
- motion parameters
- optical flow
- database
- machine learning
- motion model
- text mining
- optimization algorithm
- case based reasoning
- moving objects
- word level
- text classification
- motion segmentation
- human motion
- text retrieval
- optimization method
- point correspondences
- spatial and temporal
- relevance feedback
- intelligent systems
- knowledge representation
- keywords
- motion detection
- humanoid robot
- motion capture
- image alignment
- ai systems