In a surround view system, the image color and tone captured by multiple cameras can be different due to cameras applying auto white balance (AWB), global tone mapping (GTM) individually for each camera. The color and brightness along stitched seam location may look discontinuous among multiple cameras which impacts overall stitched image visual quality. To improve the color transition between adjacent cameras in stitching algorithm, we propose harmonization algorithm which applies before stitching to adjust multiple cameras' color and tone so that stitched image has smoother color and tone transition between adjacent cameras. Our proposed harmonization algorithm consists of AWB harmonization and GTM harmonization leveraging Image Signal Processor (ISP)'s AWB and GTM metadata statistics. Experiment result shows that our proposed algorithm outperforms global color transfer method in both visual quality and computational cost.

Wearable ultrasound imaging technology has become an emerging modality for the continuous monitoring of deep-tissue physiology, providing crucial health and disease information. Fast volumetric imaging that can provide a full spatiotemporal view of intrinsic 3D targets is desirable for interpreting internal organ dynamics. However, existing 1D ultrasound transducer arrays provide 2D images, making it challenging to overcome the trade-off between the temporal resolution and volumetric coverage. In addition, the high driving voltage limits their implementation in wearable settings. With the use of microelectromechanical system (MEMS) technology, we report an ultrasonic phased-array transducer, i.e., a 2D piezoelectric micromachined ultrasound transducer (pMUT) array, which is driven by a low voltage and is chip-compatible for fast 3D volumetric imaging. By grouping multiple pMUT cells into one single drive channel/element, we propose an innovative cell–element–array design and operation of a pMUT array that can be used to quantitatively characterize the key coupling effects between each pMUT cell, allowing 3D imaging with 5-V actuation. The pMUT array demonstrates fast volumetric imaging covering a range of 40 mm × 40 mm × 70 mm in wire phantom and vascular phantom experiments, achieving a high temporal frame rate of 11 kHz. The proposed solution offers a full volumetric view of deep-tissue disorders in a fast manner, paving the way for long-term wearable imaging technology for various organs in deep tissues.

More and more home videos have been generated with the ever growing popularity of digital cameras and camcorders. In many cases of home video, a photo, whether capturing a moment or a scene within the video, provides a complementary representation to the video. In this paper, a complete solution of video to photo is presented. The intent of the user is first derived by analyzing video motions. Then, photos are produced accordingly from the video. They can be keyframes at video highlights, panorama of the scene, or high-resolution frames. Methods and results of camera motion mining, intelligent keyframe extraction, video frame stitching and super-resolution enhancement are described.

Qian Lin, Tong Zhang, Mei Chen, Yining Deng, Pere Obrador Imaging Systems Laboratory HP Laboratories Palo Alto HPL-2004-80 April 29, 2004* home video analysis, video to photo, video printing, camera motion analysis, intelligent keyframe extraction, video panorama, super-resolution, user intent, motion miningMore and more home videos have been generated with the ever growing popularity of digital cameras and camcorders. In many cases of home video, a photo, whether capturing a moment or a scene within the video, provides a complementary representation to the video. In this paper, a complete solution of video to photo is presented. The intent of the user is first derived by analyzing video motions. Then, photos are produced accordingly from the video. They can be keyframes at video highlights, panorama of the scene, or high-resolution frames. Methods and results of camera motion mining, intelligent keyframe extraction, video frame stitching and super-resolution enhancement are described.

Abstract: this paper, acomplete solution of video to photo is presented. The intent of the user isfirst derived by analyzing video motions. Then, photos are producedaccordingly from the video. They can be keyframes at video highlights,panorama of the scene, or high-resolution frames. Methods and results ofcamera motion mining, intelligent keyframe extraction, video framestitching and super-resolution enhancement are described