Would shot detection and ball tracking be easier with stereo camera versus solo camera?

asked 2016-10-07 19:24:56 -0500

Crashalot gravatar image

updated 2016-10-10 17:29:42 -0500

Assume two camera configurations on a basketball court:

1) Solo: One smartphone on a tripod.

2) Stereo: Using stereo camera like Bumblebee (or custom one) with two cameras, each sensor/lens 12 inches apart.

The computer vision goals are: (1) track the basketball; (2) track a player; (3) detect made shots; (4) detect shot distance; and (5) detect shot angle.

Are any of these goals easier with a stereo camera (configuration #2), or are they as easily achievable with a solo camera (configuration #1)?

EDIT: This paper (http://www.ai.sri.com/~beymer/vsam/ic...) suggests a stereo camera would offer advantages over a solo camera, but the paper is also very old (1999). Is player & shot tracking better with a stereo camera, or are solo cameras equally as effective?

edit retag flag offensive close merge delete



I think neither is "easily achievable". If you are new to computer vision I would start with a single camera. Calibrate this and Identify camera pose relative the the backboard. Then maybe perform color segmentation of the ball. The size of the ball in the frame could help in identifying position (after camera orientation relative to the backboard is established). Using two phones for traditional stereo is a bad idea. You will have a hard time establishing their relative positions and synchronizing the video.

Der Luftmensch gravatar imageDer Luftmensch ( 2016-10-08 09:47:15 -0500 )edit

Sorry should have been clearer: the smartphones were meant to simplify the example, but what if you could use a stereo camera like the Bumblebee? @Der Luftmensch

Crashalot gravatar imageCrashalot ( 2016-10-09 15:38:51 -0500 )edit

I think it's inevitable that you will be disappointed with the accuracy and coverage of traditional passive stereo, particularly in structured (textureless and repetitive) environments like a gymnasium. The accuracy and coverage is much better with structured light and time-of-flight sensors, though most of these have a range too limited for your use. If and when it becomes accessible, you could try the Kinect for Xbox One which looks like it might have the range you need.

Der Luftmensch gravatar imageDer Luftmensch ( 2016-10-09 21:06:38 -0500 )edit

Thanks for the response! So you're saying a stereo camera offers no advantages over a solo camera for tracking players and shots?

Crashalot gravatar imageCrashalot ( 2016-10-09 23:08:57 -0500 )edit

Obviously there are advantages to using a stereo camera, but what I'm saying is that using stereo will present its own challenges, especially in your use case. So, don't approach it as a 'silver bullet'. Certainly you could use a multi-camera system and there are such systems in use in the NBA, but I'm guessing your time and resources will not get you to that level.

Der Luftmensch gravatar imageDer Luftmensch ( 2016-10-10 08:01:11 -0500 )edit

Thanks @Der Luftmensch for the reply! To help future readers and myself, could you kindly elaborate on what are the advantages of stereo camera? The drawbacks are added complexity in hardware and software. But for tracking players and shots, could a solo camera do everything a stereo camera could? Re the textureless issue for stereo cameras, could you solve that problem adding IR textures to stitch together the images?

Crashalot gravatar imageCrashalot ( 2016-10-10 12:58:09 -0500 )edit