Abstract: Contemporary deep learning approaches for optical flow estimation continue to face persistent challenges in model interpretability, generalization capacity, and deployment efficiency, ...

Every year, California agencies report to the state technology department on their use of "high-risk" decision systems. The seal of the State of California on the Edmund G. Brown Jr. State Office ...

UFM (Unified Flow & Matching, UniFlowMatch) is a simple, end-to-end trained transformer model that directly regresses pixel displacement images (flow) and can be applied concurrently to both optical ...

Abstract: Effective movement primitives should be capable of encoding and generating a rich repertoire of trajectories conditioned on task-defining parameters such as vision or language inputs. While ...