Pseudo Depth Meets Gaussian: A Feed-forward RGB SLAM Baseline
This work addresses the problem of slow and sensor-dependent 3D reconstruction for robotics and AR/VR applications, offering a significant speed improvement but is incremental as it builds on existing SLAM and Gaussian techniques.
The paper tackles the challenge of real-time 3D reconstruction from RGB video without depth sensors by proposing a feed-forward SLAM method using 3D Gaussian mapping and optical flow for pose prediction, achieving performance comparable to state-of-the-art while reducing tracking time by over 90%.
Incrementally recovering real-sized 3D geometry from a pose-free RGB stream is a challenging task in 3D reconstruction, requiring minimal assumptions on input data. Existing methods can be broadly categorized into end-to-end and visual SLAM-based approaches, both of which either struggle with long sequences or depend on slow test-time optimization and depth sensors. To address this, we first integrate a depth estimator into an RGB-D SLAM system, but this approach is hindered by inaccurate geometric details in predicted depth. Through further investigation, we find that 3D Gaussian mapping can effectively solve this problem. Building on this, we propose an online 3D reconstruction method using 3D Gaussian-based SLAM, combined with a feed-forward recurrent prediction module to directly infer camera pose from optical flow. This approach replaces slow test-time optimization with fast network inference, significantly improving tracking speed. Additionally, we introduce a local graph rendering technique to enhance robustness in feed-forward pose prediction. Experimental results on the Replica and TUM-RGBD datasets, along with a real-world deployment demonstration, show that our method achieves performance on par with the state-of-the-art SplaTAM, while reducing tracking time by more than 90\%.