Camera Tracks Ping Pong Ball at 1,000 fps

We have seen many vision based tracking system but this is one technology that amazes me.

Professor Masatoshi Ishikawa from the University of Tokyo has developed a pan-tilt unit for a 1,000Hz camera for tracking the ball but instead of moving the camera, 2 mirrors are used to reduce the weight of the system for it to perform such agile and high speed motion. This is explained on their site.

The following tracking is also done in the lab but with the camera mounted on the pan-tilt unit. The weight of the moving camera does matter when it concerns high speed tracking.

36 Million Faces in a Second

This camera system by Hitachi Kokusai Electric can capture, scan and recognize 36 million faces in an image in one second. This surveillance technology places it far above the competition.

Hitachi’s software is able to recognize a face with up to 30 degrees of deviation turned vertically and horizontally away from the camera, and requires faces to fill at least 40 pixels by 40 pixels for accurate recognition. Any image, whether captured on a mobile phone, handheld camera, or a video still, can be uploaded and  searched against its database for matches.

This reminds me of image scanning software used to track people in movies like “Eagle Eye”. This will enable large organisations like governments to perform large scale tracking of crowded spaces for security or marketing reasons.

facebook and face.com use another method to collect and store a database of their own. By using crowd sourcing, they are able to more accurately identify face features on the many photos online.