V-FUSE: Volumetric Depth Map Fusion with Long-Range Constraints
This work addresses depth map fusion for 3D reconstruction, offering a domain-specific solution that is incremental as it builds on existing MVS methods.
The paper tackles the problem of improving depth and confidence maps from Multi-View Stereo by introducing a learning-based fusion framework that integrates volumetric visibility constraints and reduces search space, resulting in substantial accuracy improvements in experiments.
We introduce a learning-based depth map fusion framework that accepts a set of depth and confidence maps generated by a Multi-View Stereo (MVS) algorithm as input and improves them. This is accomplished by integrating volumetric visibility constraints that encode long-range surface relationships across different views into an end-to-end trainable architecture. We also introduce a depth search window estimation sub-network trained jointly with the larger fusion sub-network to reduce the depth hypothesis search space along each ray. Our method learns to model depth consensus and violations of visibility constraints directly from the data; effectively removing the necessity of fine-tuning fusion parameters. Extensive experiments on MVS datasets show substantial improvements in the accuracy of the output fused depth and confidence maps.