packnet-sfm

Uses symmetric packing and unpacking blocks with 3D convolutions to compress detail-preserving representations, enabling high-resolution depth prediction as shown in the CVPR 2020 paper.

Trained self-supervised only on monocular videos, eliminating the need for expensive depth labeling, which is a core advantage highlighted in the framework's description.

Optimized for real-time performance using TensorRT, making it suitable for autonomous driving applications where speed is critical, as noted in the README.

Extends to non-pinhole cameras like fisheye through Neural Ray Surfaces, allowing depth estimation beyond traditional models, based on the 3DV 2020 implementation.