{\alpha}Depth: Learning Single-Pass Soft Boundary Decomposition for Stereo Conversion 事件

{\alpha}Depth: Learning Single-Pass Soft Boundary Decomposition for Stereo Conversion arXiv:2606.00386v1 Announce Type: new Abstract: Accurately modeling soft boundaries, e.g., hair and defocus blur, is a fundamental challenge in stereo conversion due to the ambiguous blending of foreground and background. Existing depth models primarily predict single-layer depth, leading to ambiguity in depth correspondence at soft boundaries. While matting techniques can capture opacity for layered modeling,