Abstract: An image is compressed according to a compression ratio selected based on a compression metric. In one embodiment, the compression metric is based on image characteristics indicative of the amount of image noise in an image, such as gain and lux. Greater compression ratios are used for image having compression metrics indicating a higher degree of noise. Because an image with higher image noise levels already has a reduced visual quality, the impact of higher compression is less significant as compared to the impact of compression on images having low image noise. In an embodiment, an image is divided into a number of regions, for each of which a compression metric and corresponding compression ratio is determined, so that regions of the image having high image noise may be compressed more than regions having low image noise.

Abstract: An image is compressed according to a compression ratio selected based on a compression metric. In one embodiment, the compression metric is based on image characteristics indicative of the amount of image noise in an image, such as gain and lux. Greater compression ratios are used for image having compression metrics indicating a higher degree of noise. Because an image with higher image noise levels already has a reduced visual quality, the impact of higher compression is less significant as compared to the impact of compression on images having low image noise. In an embodiment, an image is divided into a number of regions, for each of which a compression metric and corresponding compression ratio is determined, so that regions of the image having high image noise may be compressed more than regions having low image noise.

Abstract: Systems, methods, and computer readable media for stabilizing video frames based on information from a motion sensor are described. In general, digital video stabilization techniques are disclosed for generating and applying image-specific transformations to individual frames (images) in a video sequence after, rather than before, the image has been captured. The transformations may be used to counter-balance or compensate for unwanted jitter occurring during video capture due to, for example, a person's hand shaking.

Abstract: Systems, methods, and computer readable media for stabilizing video frames based on information from a motion sensor are described. In general, digital video stabilization techniques are disclosed for generating and applying image-specific transformations to individual frames (images) in a video sequence after, rather than before, the image has been captured. The transformations may be used to counter-balance or compensate for unwanted jitter occurring during video capture due to, for example, a person's hand shaking.

Abstract: Systems, methods and a computer readable mediums for reducing run-time bandwidth requirements within digital image capture devices are disclosed herein. As the resolution of digital image capture devices increase, the bandwidth needed to support this increased resolution is becoming increasingly difficult to support. To reduce the problems associated with run-time bandwidth requirements, it may be beneficial to obtain full-resolution and display-resolution images at the same time from the image capture circuitry—writing both to memory at virtually the same time. The display resolution image may then be delivered to a display unit directly without the need for additional memory operations on the full-resolution image. Only when a user indicates they wish to capture an image need additional memory operations on the full-resolution image be performed.