Amazon Technologies, Inc.

Maxim Arap

Chun-Hao Liu

Sheng Liu

Sheik Dawood Beer Mohideen

Camera positions can be recovered for video frames. Keypoint tracks can be generated for a first video frame and subsequent video frames. Metric depth can be determined for the video frames. An image databased can be generated using the keypoint tracks and the metric depth. The image database can include the video frames. A first video frame having a highest number of keypoint tracks can be identified. A first projection into a three-dimensional model can be generated for the first video frame and a first matching video frame. A camera position can be generated using the first projection. A second video frame having a next-highest number of keypoint tracks can be identified. A second projection into a three-dimensional model can be generated for the second video frame and a second matching video frame. A camera position can be generated using the second projection.

Recovering camera position from content file

At least part of a visual pattern may be detected, in an image of a video, at a location within the image. An occlusion determination may be performed corresponding to the at least part of the visual pattern. At least part of a virtual asset may be inserted, based at least in part on the occlusion determination, into the image at the location. An appearance of the at least part of the virtual asset within the image may be adjusted based at least in part on an appearance of the at least part of the visual pattern within the image.

Growing community of inventors

San Jose, CA, United States of America

Maxim Arap