Radiant Triangle Soup with Soft Connectivity Forces for 3D Reconstruction and Novel View Synthesis

In this work, we introduce an inference-time optimization framework utilizing triangles to represent the geometry and appearance of the scene. More specifically, we develop a scene optimization algorithm for triangle soup, a collection of disconnected semi-transparent triangle primitives. Compared to the current most-widely used primitives for 3D scene representation, namely Gaussian splats, triangles allow for more expressive color interpolation, and benefit from a large algorithmic infrastructure for downstream tasks. Triangles, unlike full-rank Gaussian kernels, naturally combine to form surfaces. We formulate connectivity forces between triangles during optimization, encouraging explicit, but soft, surface continuity in 3D. We perform experiments on a representative 3D reconstruction dataset and show competitive photometric and geometric results.

V-FUSE: Volumetric Depth Map Fusion with Long-Range Constraints

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.

Real-Time Dense 3D Mapping of Underwater Environments

This paper addresses real-time dense 3D reconstruction for a resource-constrained Autonomous Underwater Vehicle (AUV). Underwater vision-guided operations are among the most challenging as they combine 3D motion in the presence of external forces, limited visibility, and absence of global positioning. Obstacle avoidance and effective path planning require online dense reconstructions of the environment. Autonomous operation is central to environmental monitoring, marine archaeology, resource utilization, and underwater cave exploration. To address this problem, we propose to use SVIn2, a robust VIO method, together with a real-time 3D reconstruction pipeline. We provide extensive evaluation on four challenging underwater datasets. Our pipeline produces comparable reconstruction with that of COLMAP, the state-of-the-art offline 3D reconstruction method, at high frame rates on a single CPU.