Court Reconstruction for Camera Calibration in Broadcast Basketball Videos

Abstract

We introduce a technique of calibrating camera motions in basketball videos. Our method particularly transforms player positions to standard basketball court coordinates and enables applications such as tactical analysis and semantic basketball video retrieval. To achieve a robust calibration, we reconstruct the panoramic basketball court from a video, followed by warping the panoramic court to a standard one. As opposed to previous approaches, which individually detect the court lines and corners of each video frame, our technique considers all video frames simultaneously to achieve calibration; hence, it is robust to illumination changes and player occlusions. To demonstrate the feasibility of our technique, we present a stroke-based system that allows users to retrieve basketball videos. Our system tracks player trajectories from broadcast basketball videos. It then rectifies the trajectories to a standard basketball court by using our camera calibration method. Consequently, users can apply stroke queries to indicate how the players move in gameplay during retrieval. The main advantage of this interface is an explicit query of basketball videos so that unwanted outcomes can be prevented. We show the results in Figs. 1, 7, 9, 10 and our accompanying video to exhibit the feasibility of our technique.

A Two-Point Method for PTZ Camera Calibration in Sports

“…However, these parameters may change from one game to another. In that case, fully automatic methods [37,16] can be used to estimate the camera base parameters before using our method. Second, our camera model does not consider lens distortion which may be significant in some games.…”

“…SIFT [22]) from a reference image and the new image. These approaches have been extended to automatic reference generation [37,16] and line/ellipse matching [9,13,21]. However, human annotation is still needed to get ground truth data and a small set of reference images.…”

Sports Camera Calibration via Synthetic Data

“…The current trend is using highly/fully automatic methods to deal with large-scale data. For example, Wen et al [2] first reconstructed a panoramic court image from a basketball video. Then, they warped the panoramic court to the court template.…”

“…The images consist of different views of the field with different grass textures, lighting patterns and heavy shadows. We pre-process the training set to obtain following camera configurations: camera location distribution N (µ, σ 2 ) (µ ≈ [52, −45, 17] T and σ ≈ ±[2,9,3] T meters); pan, tilt and focal length ranges ([−35 • , 35 • ], [−15 • , −5 • ]…”

Collaboration calibration and three-dimensional localization in multi-view system

“…Many researches focus on the accuracy of the parameters and localization results. 6,[13][14][15] However, the collaboration calibration of multi-view cameras is convenient and efficient at the deployment phase of indoor surveillance. It can reduce the heavy work of multi-camera calibration in visual sensor networks.…”