We propose a unified point cloud video self-supervised learning framewor...
Recently, the community has made tremendous progress in developing effec...
The commonly adopted detect-then-match approach to registration finds
di...
RGB-T semantic segmentation has been widely adopted to handle hard scene...
Self-supervised learning can extract representations of good quality fro...
In this report, we summarize the first NTIRE challenge on light field (L...
Single-frame infrared small target (SIRST) detection aims at separating ...
In this paper, we present a new method for the multiview registration of...
Given a 3D object, kinematic motion prediction aims to identify the mobi...
Pseudo-Labeling has emerged as a simple yet effective technique for
semi...
Exploiting spatial-angular correlation is crucial to light field (LF) im...
Recently, unsupervised domain adaptation in satellite pose estimation ha...
Space-based infrared tiny ship detection aims at separating tiny ships f...
Recent years have witnessed the great advances of deep neural networks (...
Visual speech, referring to the visual domain of speech, has attracted
i...
With the development of the 3D data acquisition facilities, the increasi...
In this paper, we summarize the 1st NTIRE challenge on stereo image
supe...
Learning-based multi-view stereo (MVS) has by far centered around 3D
con...
Deep models trained on source domain lack generalization when evaluated ...
We study the problem of efficient object detection of 3D LiDAR point clo...
It is well known that the passive stereo system cannot adapt well to wea...
We study the problem of attribute compression for large-scale unstructur...
Matching cost construction is a key step in light field (LF) depth
estim...
Light field (LF) cameras record both intensity and directions of light r...
We study the problem of extracting accurate correspondences for point cl...
Learning dense point-wise semantics from unstructured 3D point clouds wi...
Weakly supervised learning can help local feature methods to overcome th...
Satellite video cameras can provide continuous observation for a large-s...
Effective learning of spatial-temporal information within a point cloud
...
Camera arrays provide spatial and angular information within a single
sn...
Domain adaptation is critical for success when confronting with the lack...
Single-frame infrared small target (SIRST) detection aims at separating ...
Most existing CNN-based super-resolution (SR) methods are developed base...
Scene classification, aiming at classifying a scene image to one of the
...
The real-time performance of the stereo matching network is important fo...
Extracting robust and general 3D local features is key to downstream tas...
Although recent years have witnessed the great advances in stereo image
...
A main challenge for tasks on panorama lies in the distortion of objects...
Graph neural networks (GNNs) have recently grown in popularity in the fi...
Stereo image pairs encode 3D scene cues into stereo correspondences betw...
To have a better understanding and usage of Convolution Neural Networks
...
Light field (LF) cameras can record scenes from multiple perspectives, a...
Pseudo-LiDAR point cloud interpolation is a novel and challenging task i...
Current CNN-based super-resolution (SR) methods process all locations eq...
Keypoint detection and description is fundamental yet important in many
...
Recently, the performance of single image super-resolution (SR) has been...
The spatio-temporal information among video sequences is significant for...
Video super-resolution (SR) aims at generating a sequence of high-resolu...
Point cloud learning has lately attracted increasing attention due to it...
Light field (LF) cameras record both intensity and directions of light r...