Establishing reliable correspondences between two views is one of the most important components of various vision tasks. This paper proposes a novel sparse-to-local-dense (S2LD) matching method to conduct fully differentiable correspondence estimation with the prior from epipolar geometry. The sparse-to-local-dense matching asymmetrically establishes correspondences with consistent sub-pixel coordinates while reducing the computation of matching. The salient features are explicitly located, and the description is conditioned on both views with the global receptive field provided by the attention mechanism. The correspondences are progressively established in multiple levels to reduce the underlying re-projection error. We further propose a 3D noise-aware regularizer with differentiable triangulation. Additional guidance from 3D space is encoded by the regularizer in training to handle the supervision noise caused by the errors in camera poses and depth maps. The proposed method demonstrates outstanding matching accuracy and geometric estimation capability on multiple datasets and tasks.
1 June 2021 in IEEE/ASME Transactions on Mechatronics
The indirect VSLAM is widely used in robot localization and navigation thanks to its potential to achieve high localization accuracy with the local feature observations. However, the existing local features are subject to drift and mismatches under various visual conditions, which causes a degrading in localization accuracy and tracking loss. This paper proposes a quantized self-supervised local feature for the indirect VSLAM to handle the environmental interference in robot localization tasks. A joint feature detection and description network is built in a lightweight manner to extract local features in real time. The network is iteratively trained by a self-supervised learning strategy, and the extracted local features are quantized by an orthogonal transformation for efficiency. We utilize frame-wise matching in Hamming space and bundle adjustment to establish a parallel indirect VSLAM. The proposed VSLAM demonstrates outstanding localization accuracy and tracking stability in the evaluation on multiple datasets and robustness in real-world experiments with the Realsense D435 RGB-D sensor. The efficiency experiment on Jetson TX2 indicates that the quantized self-supervised local feature is suitable for feature-based tasks on edge computing platforms.
31 July 2023 in IEEE Transactions on Circuits and Systems for Video Technology
Implicit scene representations have recently shown promising results in photo-realistic 3D reconstruction and view synthesis based on calibrated views. However, their applications face several challenges, including unknown camera pose, boundary ambiguity, and observation noise. This paper proposes a novel online scene representation method that simultaneously learns to represent the target scene and estimates the camera poses from an RGB-D stream. An implicit scene representation function built with scale-encoded cascaded grids is proposed to represent scenes online from incremental observations. This implicit function is optimized in a reparameterized domain that provides defined boundaries. In this reparameterized domain, the cascaded grids are progressively distilled under geometric and photometric supervision to improve their model capacity and geometry accuracy. A radiance field deblurring module based on the physical imaging process is further proposed to restore a photo-realistic reconstruction against camera motion blur, which is the main component of the observation noise. The proposed method can produce sharp and photo-realistic representations of scenes under various shooting conditions without known camera poses. Experiments on multiple datasets have demonstrated the effectiveness of the proposed method in improving view synthesis and camera tracking results for online scene representation tasks.