Search Results for author:
Found 3 papers, 0 papers with code
Scene-Aware Feature Matching
Current feature matching methods focus on point-level matching, pursuing better representation learning of individual features, but lacking further understanding of the scene.
ParaFormer: Parallel Attention Transformer for Efficient Feature Matching
Heavy computation is a bottleneck limiting deep-learningbased feature matching algorithms to be applied in many realtime applications.
Self-Supervision and Spatial-Sequential Attention Based Loss for Multi-Person Pose Estimation
To solve these problems, this paper proposes (i) a new loss organization method which uses self-supervised heatmaps to reduce prediction contradictions and spatial-sequential attention to enhance networks' features extraction; (ii) a new combination of predictions composed by heatmaps, Part Affinity Fields (PAFs) and our block-inside offsets to fix pixel-level joints positions and further demonstrates the effectiveness of proposed loss function.
