Abstract: A drivable implicit three-dimensional human body representation method, which is used for performing dynamic reconstruction by means of optimizing a three-dimensional representation of a drivable model from an input multi-view video.

Abstract: The present application discloses a LiDAR and a movable device. The LiDAR includes an emission module, a receiving module, a beam splitter and a scanning module. The receiving lens meets 40 mm?fRX?80 mm, where fRX is the effective focal length of the receiving lens. The beam splitter includes a light transmitting portion and a reflecting portion, the reflecting portion being disposed at the periphery of the light transmitting portion. The scanning module is configured to emit the detected light to the target object, and to transmit the echo light to the receiving module.

Abstract: This application discloses an off-axis optical system and a LiDAR. The off-axis optical system includes a transmitting module, a receiving module, and a deflecting element. The transmitting module and the receiving module are arranged along a first direction perpendicular to the optical axis of the receiving module. The deflecting element is positioned on a light-incident side of the receiving module, with a gap existing along the first direction between the deflecting element and the optical axis of the receiving module. At least part of echo beams, which are formed by reflection of scanning beams emitted from the transmitting module off a target, are refracted by the deflecting element before reaching the receiving module.

Abstract: Disclosed is a human motion capture method based on unsynchronized videos, which can effectively recover the 3D motion of a target person through multiple unsynchronized videos of the person. In order to utilize multiple unsynchronized videos, the present disclosure provides a video synchronization and motion reconstruction method. The present disclosure is implemented in the following steps: synchronizing multiple videos based on a 3D human pose; performing motion reconstruction based on synchronized videos, modeling the motion difference across different viewpoints by using the low-rank constraint to realize high-precision human motion capture from the plurality of unsynchronized videos. According to the present disclosure, more accurate motion capture is carried out by using multiple unsynchronized videos.

Abstract: A drivable implicit three-dimensional human body representation method, which is used for performing dynamic reconstruction by means of optimizing a three-dimensional representation of a drivable model from an input multi-view video.

Abstract: Disclosed is a human motion capture method based on unsynchorized videos, which can effectively recover the 3D motion of a target person through multiple unsynchorized videos of the person. In order to utilize multiple unsynchorized videos, the present disclosure provides a video synchronization and motion reconstruction method. The present disclosure is implemented in the following steps: synchronizing multiple videos based on a 3D human pose; performing motion reconstruction based on synchronized videos, modeling the motion difference across different viewpoints by using the low-rank constraint to realize high-precision human motion capture from the plurality of unsynchorized videos. According to the present disclosure, more accurate motion capture is carried out by using multiple unsynchorized videos.