S²M²: Scalable Stereo Matching Model for Reliable Depth Estimation
Published in IEEE/CVF International Conference on Computer Vision (ICCV), 2025
S²M² revisits global matching with a scalable multi-resolution transformer and a loss that encourages reliable disparity and confidence estimation.
Recommended citation: Junhong Min, Youngpil Jeon, Jimin Kim, and Minyong Choi. "S²M²: Scalable Stereo Matching Model for Reliable Depth Estimation." ICCV, 2025.
Download Paper