Back to results
US 11,960,082 B2Granted
Virtual reality system and object detection method applicable to virtual reality system
AssigneeHTC Corporation
InventorsJyun-Jhong Lin, Chun-Kai Huang, Heng-Li Hsieh, Yun-Jung Chang
Life of this patent
Estimating the term…
Abstract
An object detection method is suitable for a virtual reality system. The object detection method includes a plurality of first cameras of a head-mounted display (HMD) to capture a plurality of first frames. A plurality of second frames are captured through a plurality of second cameras in a tracker, wherein, the object detector searches for the object position in the first frames and the second frames.
Claims14
A virtual reality system, comprising:
- a head-mounted display (HMD), configured to be fixed on a head of a user, and comprising:
- a plurality of first cameras, configured to capture an environment to obtain a plurality of first frames, wherein the environment comprises an object;
- a storage device;
- and
- a processor, configured to access an object detector stored in the storage device to execute the object detector, wherein the processor inputs the first frames and a first map data of each of the first frames into the object detector;
- and
- a tracker, configured to be fixed on a limb of the user, and comprising:
- a plurality of second cameras, configured to capture the environment to obtain a plurality, of second frames;
- wherein the tracker transmits the second frames and a second map data of each of the second frames to the object detector, and the object detector updates an environment coverage frame in response to a determination that the first map data and the second map data are different from existing data at a same location, determines a three-dimensional intersection area of the first frames and the second frames, wherein the intersection area comprises a plurality of target feature points corresponding to the object, and searches for an object position of the object in the first frames and the second frames according to the plurality of target feature points.
6 more elements
A object detection method, suitable for a virtual reality system, wherein the object detection method comprises:
- capturing an environment to obtain a plurality of first frames through a plurality of first cameras of a head-mounted display (MMD), wherein the head-mounted display (HMD) is configured to be fixed on a head of a user, and the environment comprises an object;
- executing the following steps using a processor of the head-mounted display:
- inputting the first frames and a first map data of each of the first frames into an object detector;
- and
- capturing the environment to obtain a plurality of second frames using a plurality of second cameras in a tracker, wherein the tracker is configured to be fixed on a limb of the user;
- wherein the tracker transmits the second frames and a second map data of each of the second frames to the object detector, and the object detector updates an environment coverage frame in response to a determination that the first map data and the second map data are different from existing data at a same location, determines a three-dimensional intersection area of the first frames and the second frames, wherein the intersection area comprises a plurality of target feature points corresponding to the object, and searches for an object position of the object in the first frames and the second frames according to the plurality of target feature points.
3 more elements
Dependencies are inferred from the claim text.