Consistent Depth is a research project developed by Facebook Research that presents an algorithm for reconstructing dense and geometrically consistent depth information for all pixels in a monocular video. The system builds upon traditional structure-from-motion (SfM) techniques to provide geometric constraints while integrating a convolutional neural network trained for single-image depth estimation. During inference, the model fine-tunes itself to align with the geometric constraints of a specific input video, ensuring stable and realistic depth maps even in less-constrained regions. This approach achieves improved geometric consistency and visual stability compared to prior monocular reconstruction methods. The project can process challenging hand-held video footage, including those with moderate dynamic motion, making it practical for real-world usage.
Features
- Reconstructs dense, geometrically consistent depth from monocular videos
- Integrates structure-from-motion with deep learning-based depth priors
- Supports test-time fine-tuning for video-specific geometric constraints
- Provides higher accuracy and consistency than classical monocular reconstruction methods
- Compatible with dynamic hand-held videos and real-world camera motion
- Enables scene reconstruction and video visual effects through improved depth estimation
Project Samples
