Processing of video data to compensate for unintended camera motion between acquired image frames
View Patent ↗Video data that includes a plurality of images acquired over time are processed in an apparatus immediately after acquiring the video data, or during some later post-processing of the video data, to reduce the effects of unintended motion in the acquired images. Coarse and fine compensation for motion is performed on video frames that correspond to the acquired images. A coarse compensation value is employed to compress each video frame that is both coarse and fine motion compensated.
1. A method for enabling an apparatus to generate a video output, comprising:
employing a lens and a sensor to acquire video data that includes a plurality of images acquired over time, and performing actions on each acquired images, including:
determining a first global motion vector for a current image based on detected coarse motion between a low resolution version of the current image and at least one of a plurality of earlier acquired images having a resolution that is relatively lower than the current image;
cropping an intermediate video frame from the current image, wherein the intermediate video frame is repositioned within a region of the current image to compensate for the detected coarse motion based on the determined first global motion vector;
determining a second global motion vector for the intermediate video frame based on detected fine motion between the intermediate video frame and at least one earlier intermediate video frame;
cropping a video frame from the intermediate video frame, wherein the video frame is repositioned within a region of the intermediate video frame to compensate for the detected fine motion based on the determined second global motion vector; and
compressing the video frame based on the determined first global motion vector; and
generating a video output that includes a plurality of the compressed video frames over time.
2. The method of claim 1 , further comprising employing a low resolution version of earlier acquired images to detect coarse motion and determine the first global motion vector for the current image.
3. An integrated circuit for enabling an apparatus to generate a video output, comprising:
a memory component that stores at least instructions; and
a processing component that executes instructions to enable a plurality of actions, including:
enabling a lens and a sensor to acquire video data that includes a plurality of images acquired over time, and performing actions on each acquired image, comprising:
determining a first global motion vector for a current image based on detected coarse motion between a low resolution version of the current image and at least one of a plurality of earlier acquired images having a resolution that is relatively lower than the current image;
cropping an intermediate video frame from the current image, wherein the intermediate video frame is repositioned within a region Of the current image to compensate for the detected coarse motion based on the determined first global motion vector;
determining a second global motion vector for the intermediate video frame based on detected fine motion between the intermediate video frame and at least one earlier intermediate video frame;
cropping a video frame from the intermediate video frame, wherein the video frame is repositioned within a region of the intermediate video frame to compensate for the detected fine motion based on the determined second global motion vector; and
compressing the video frame based on the determined first global motion vector; and
generating a video output that includes a plurality of the compressed video frames over time.
4. The integrated circuit of claim 3 , wherein the actions further comprise employing a low resolution version of earlier acquired images to detect coarse motion and determine the first global motion vector for the current image.
5. A system for generating a video output, comprising:
a lens;
a sensor for acquiring an image projected by the lens onto the sensor;
a memory component for storing at least instructions; and
a processing component that executes the instructions to enable a plurality of actions, including:
enabling the lens and the sensor to acquire video data that includes a plurality of images acquired over time, and performing actions on each acquired image, comprising:
determining a first global motion vector for a current image based on detected coarse motion between a low resolution version of the current image and at least one of a plurality of earlier acquired images having a resolution that is relatively lower than the current image;
cropping an intermediate video flame from the current image, wherein the intermediate video frame is repositioned within a region of the current image to compensate for the detected coarse motion based on the determined first global motion vector;
determining a second global motion vector for the intermediate video frame based on detected fine motion between the intermediate video frame and at least one earlier intermediate video frame;
cropping a video frame from the intermediate video frame, wherein the video frame is repositioned within a region of the intermediate video frame to compensate for the detected fine motion based on the determined second global motion vector; and
compressing the video frame based on the determined first global motion vector; and
generating the video output that includes a plurality of the compressed video frames over time.
6. The system of claim 5 , wherein the actions further comprise employing a low resolution version of earlier acquired images to detect coarse motion and determine the first global motion vector for the current image.
7. A processor readable non-transitory storage media that is arranged to store executable instructions, wherein an execution of the stored instructions enables a plurality of actions for generating a video output that is motion compensated, the plurality of actions comprising:
enabling a lens and a sensor to acquire video data that includes a plurality of images acquired over time, and performing actions on each acquired image, comprising:
determining a first global motion vector for a current image based on detected coarse motion between a low resolution version of the current image and at least one of a plurality of earlier acquired images having a resolution that is relatively lower than the current image;
cropping an intermediate video frame from the current image, wherein the intermediate video frame is repositioned within a region of the current image to compensate for the detected coarse motion based on the determined first global motion vector;
determining a second global motion vector for the intermediate video frame based on detected tine motion between the intermediate video frame and at least one earlier intermediate video frame;
cropping a video frame from the intermediate video frame, wherein the video frame is repositioned within a region of the intermediate video frame to compensate for the detected fine motion based on the determined second global motion vector; and
compressing the video frame based on the determined first global motion vector; and
generating a video output that includes a plurality of the compressed video frames over time.
8. The processor readable non-transitory storage media of claim 7 , wherein the actions further comprise a low resolution version of earlier acquired images to detect coarse motion and determine the first global motion vector for the current image.