Systems and methods for video compressive sensing using a rotating mask
In an example embodiment, a motion capturing device includes a mask configured to rotate during capture of a scene by the motion capturing device and modulate the scene; and a motor coupled to the mask and configured to rotate the mask during the capturing of the scene.
1. A motion capturing device comprising:
a mask configured to rotate during capture of a scene by the motion capturing device and modulate the scene; and
a motor coupled to the mask and configured to rotate the mask during the capturing of the scene,
wherein the mask is coded with a random pattern.
2. The motion capturing device of claim 1 , wherein the motor is configured to rotate the mask such that the mask completes one revolution every time a frame of the scene is captured by the motion capturing device.
3. The motion capturing device of claim 2 , wherein the motor is configured to rotate the mask for a number of times every time a frame of the scene is captured by the motion capturing device, each time the motor is configured to rotate the mask for a fixed degree of rotation.
4. The motion capturing device of claim 3 , wherein the fixed degree of rotation is based on a number of high speed frames of the scene to be reconstructed from the modulated scene.
5. The motion capturing device of claim 4 , wherein the fixed degree of rotation is given by 360 degrees divided by the number of high speed frames.
6. The motion capturing device of claim 3 , wherein
the scene is a high speed motion scene, and
the motion capturing device is configured to capture the scene at a rate of frames per unit of time that is lower relative to the high speed motion of the scene.
7. The motion capturing device of claim 1 , wherein the random pattern is a binary pattern that either allows for a transmission of light or a blocking thereof.
8. The motion capturing device of claim 1 , wherein the random pattern is a grayscale pattern according to which light is transmitted in proportion to a local transmittance ratio of the mask.
9. The motion age capturing device of claim 1 , wherein the rotation of the mask is an angular rotation about a central axis passing through components of the motion capturing device.
10. The motion capturing device of claim 1 , wherein during the capture of the scene, the mask does not shift vertically or horizontally.
11. The motion capturing device of claim 1 , wherein the motor is a vibrating motor.
12. The motion capturing device of claim 1 , further comprising:
a lens configured to project light on to the scene;
a relay lens; and
a charge-coupled device, wherein
the mask is between the lens and the relay lens.
13. The motion capturing device of claim 1 , further comprising:
a lens configured to project light on to the scene;
a relay lens; and
a charge-coupled device, wherein
the mask is between the relay lens and the charge-coupled device.
14. An electronic mobile device, comprising:
a display; and
the motion capturing device of claim 1 , the motion capturing device configured to be controlled via the display.
15. A device comprising:
a memory having computer-readable instructions stored therein; and
a processor configured to execute the computer-readable instructions to,
control a motion capturing device to capture a scene,
receive the captured scene,
obtain a plurality of frames corresponding to the scene based on the captured scene and a plurality of filters, each of the plurality of filters corresponding to a rotated version of a mask in the motion capturing device, the mask being configured to rotate during the capturing of the scene, and
reconstruct the scene based on the plurality of frames,
wherein the mask is coded with a random pattern.
16. The device of claim 15 , wherein the scene is a high speed motion scene formed of the plurality of frames.
17. The device of claim 16 , wherein the motion capturing device is configured to capture the scene at a lower rate of frames per unit of time relative to the high speed motion of the scene.
18. The device of claim 15 , wherein the processor is configured to execute the computer-readable instructions to obtain each of the plurality of frames by,
transforming the received captured scene into a first vector of pixel values, and
for each pixel value, determine a second vector of pixel values that minimizes a difference between the pixel value and a product of the second vector of pixel values and a corresponding one of the plurality of filters, the second vector of pixel values corresponding to one of the plurality of frames.
19. The device of claim 18 , wherein the product is an element-wise product of the second vector of pixel values and the corresponding one of the plurality of filters.