Disney Enterprises, Inc.

Jianhui Chen

George Peter Carr

Yisong Yue

The disclosure provides an approach for mimicking human camera operation with an autonomous camera system. In one embodiment, camera planning is formulated as a supervised regression problem in which an automatic broadcasting application receives one video input captured by a human-operated camera and another video input captured by a stationary camera with a wider field of view. The automatic broadcasting application extracts feature vectors and pan-tilt-zoom states from the stationary camera and the human-operated camera, respectively, and learns a regressor which takes as input such feature vectors and outputs pan-tilt-zoom settings predictive of what the human camera operator would choose. The automatic broadcasting application may then apply the learned regressor on newly captured video to obtain planned pan-tilt-zoom settings and control an autonomous camera to achieve the planned settings to record videos which resemble the work of a human operator in similar situations.

Method and system for mimicking human camera operation

The disclosure provides an approach for predicting trajectories for real-time capture of video and object tracking, while adhering to smoothness constraints so that predictions are not excessively jittery. In one embodiment, a temporally consistent search and learn (TC-SEARN) algorithm is applied to train a regressor for camera planning. A automatic broadcasting application first receives video input captured by a human-operated camera and another video input captured by a stationary camera with a wide field of view. The automatic broadcasting application extracts feature vectors and pan-tilt-zoom states from the stationary camera input and human-operated camera input, respectively. The automatic broadcasting application further applies the TC-SEARN algorithm to learn a sequential regressor for predicting camera trajectories, based on the extracted feature vectors and pan-tilt-zoom states. The TC-SEARN algorithm itself is able to learn the regressor using a loss function which enables decision trees to reason about spatiotemporal smoothness via an autoregressive function.

Growing community of inventors

Burbank, CA, United States of America

Jianhui Chen