Method for creating automatic cinemagraphs on an imagine device
View Patent ↗Embodiments are directed towards automatically creating a cinemagraph on at least an imaging device, where the cinemagraph can be created without additional user interaction beyond capturing an initial sequence of images, and indicating that a cinemagraph is to be created from the sequence. Automatic creation of the cinemagraph includes selecting an anchor frame within the sequence and aligning the other frames to the anchor frame. Detection and segmentation of moving objects within the sequence, with respect to the anchor frame are performed. A mask is generated and refined. Segmentation of masks are then unified, and combined with a background from the anchor frame to generate an animated sequence.
1. A processor based method operating within an image capture device, comprising:
employing one or more image sensor devices to capture a sequence of image frames, the sequence having at least some motion and some non-motion content within a captured scene; and
automatically performing actions on the captured sequence absent user input, the actions including:
selecting from within the sequence of image frames an anchor frame;
compensating for alignment inaccuracies within the sequence via registering each of the other image frames in the sequence to the anchor frame;
detecting local motion between the registered image frames and the anchor frame;
segmenting the detected local motion region in each image frame;
refining the segmented local motion region by determining a force working on a boundary pixel of the segmented local motion region and moving the local motion region boundary depending on the value of the force; and
replacing pixels in each image frame, excluding pixels associated with the detected local motion region, with corresponding pixels from the anchor frame to generate a cinemagraph.
2. The method of claim 1 , wherein capturing the sequence of image frames further comprises automatically selecting by the image capture device at least one of a focal parameter or a capture speed.
3. The method of claim 1 , wherein selecting the anchor frame comprises selecting the anchor frame as a first frame in the sequence of the images or a last frame.
4. The method of claim 1 , wherein registering each of the other image frames to the anchor frame further comprises employing a mean-normalized sum of absolute differences to perform a motion search, and using a result to generate a transformation that is applied to each of the other image frames.
5. The method of claim 1 , wherein segmenting the detected local motion further comprises detecting motion areas within the sequence by thresholding an image of normalized distances and then refining the segmentation using active contouring.
6. The method of claim 1 , wherein detecting local motion further comprises selecting a motion area based on having a maximal overlap with a mask that is determined from a previous image frame in the sequence of image frames.
7. The method of claim 1 , wherein segmenting the detected local motion region further comprises examining each pixel identified as a boundary pixel to the local motion region to assign the pixel to outside of the local motion region or the inside of the local motion region based on a value of a computed force working on the boundary pixel.
8. An image system, comprising:
an image sensor device used to capture a sequence of image frames; and
one or more processors that automatically perform actions on the captured sequence absent user input, the actions including:
selecting from within the sequence of image frames an anchor frame;
compensating for alignment inaccuracies within the sequence via registering each of the other image frames in the sequence to the anchor frame;
detecting local motion between the registered image frames and the anchor frame;
segmenting the detected local motion region in each image frame;
refining the segmented local motion region by determining a force working on a boundary pixel of the segmented local motion region and moving the local motion region boundary depending on the value of the force; and
replacing pixels in each image frame, excluding pixels associated with the detected local motion region, with corresponding pixels from the anchor frame to generate a cinemagraph.
9. The image system of claim 8 , wherein capturing the sequence of image frames further comprises automatically selecting by the image capture device at least one of a focal parameter or a capture speed.
10. The image system of claim 8 , wherein the actions further comprise adding to an end of the cinemagraph a reverse sequence of frames in the cinemagraph sequence.
11. The image system of claim 8 , wherein registering each of the other image frames to the anchor frame further comprises employing a mean-normalized sum of absolute differences to perform a motion search, and using a result to generate a transformation that is applied to each of the other image frames.
12. The image system of claim 8 , wherein segmenting the detected local motion further comprises detecting motion areas within the sequence by thresholding an image of normalized distances and then refining the segmentation using active contouring.
13. The image system of claim 8 , wherein detecting local motion further comprises selecting a motion area based on having a maximal overlap with a mask that is determined from a previous image frame in the sequence of image frames.
14. The image system of claim 8 , wherein segmenting the detected local motion region further comprises examining each pixel identified as a boundary pixel to the local motion region to assign the pixel to outside of the local motion region or the inside of the local motion region based on a value of a computed force working on the boundary pixel.
15. An apparatus for managing image data, comprising:
one or more physical components that capture a sequence of images; and
one or more circuit units that automatically perform actions on the captured sequence absent user input, the actions including:
selecting from within the sequence of image frames an anchor frame;
compensating for alignment inaccuracies within the sequence via registering each of the other image frames in the sequence to the anchor frame;
detecting local motion between the registered image frames and the anchor frame;
segmenting the detected local motion region in each image frame;
refining the segmented local motion region by determining a force working on a boundary pixel of the segmented local motion region and moving the local motion region boundary depending on the value of the force; and
replacing pixels in each image frame, excluding pixels associated with the detected local motion region, with corresponding pixels from the anchor frame to generate a cinemagraph.
16. The apparatus of claim 15 , wherein capturing the sequence of image frames further comprises automatically selecting by the image capture device at least one of a focal parameter or a capture speed.
17. The apparatus of claim 15 , wherein registering each of the other image frames to the anchor frame further comprises employing a mean-normalized sum of absolute differences to perform a motion search, and using a result to generate a transformation that is applied to each of the other image frames.
18. The apparatus of claim 15 , wherein segmenting the detected local motion further comprises detecting motion areas within the sequence by thresholding an image of normalized distances and then refining the segmentation using active contouring.
19. The apparatus of claim 15 , wherein detecting local motion further comprises selecting a motion area based on having a maximal overlap with a mask that is determined from a previous image frame in the sequence of image frames.
20. The apparatus of claim 15 , wherein the actions further comprise adding to an end of the cinemagraph a reverse sequence of frames in the cinemagraph sequence.