Systems and methods for detecting motion during 3D data reconstruction

A computerized method for detecting movement of an object in a scene, the method comprising:

• accessing a first set of images and a second set of images of a scene over time;
• generating, based on the first set of images, a first temporal pixel image comprising a first set of temporal pixels, wherein each temporal pixel in the first set of temporal pixels comprises a set of pixel values at an associated position from each image of the first set of images;
• generating, based on the second set of images, a second temporal pixel image comprising a second set of temporal pixels, wherein each temporal pixel in the second set of temporal pixels comprises a set of pixel values at an associated position from each image of the second set of images;
• determining one or more derived values based on values of the temporal pixels in the first temporal pixel image, the second temporal pixel image, or both;
• determining, based on the first temporal pixel image and the second temporal pixel image, a correspondence density indicative of a relationship between a number of a set of identified correspondences between image points of the first set of images and image points of the second set of images and a number of possible correspondences;
• determining, based on the one or more derived values and the correspondence density, an indication of whether there is a likelihood of motion of an object in the scene;
• and
• based on determining that there is a likelihood of motion of the object in the scene, determine a position of the object in the scene at a future time subsequent to a time related to when the first set of images and the second set of images were captured.

At least one non-transitory computer-readable storage medium storing processor-executable instructions that, when executed by at least one computer hardware processor, cause the at least one computer hardware processor to perform the acts of:

• accessing a first set of images and a second set of images of a scene over time;
• generating, based on the first set of images, a first temporal pixel image comprising a first set of temporal pixels, wherein each temporal pixel in the first set of temporal pixels comprises a set of pixel values at an associated position from each image of the first set of images;
• generating, based on the second set of images, a second temporal pixel image comprising a second set of temporal pixels, wherein each temporal pixel in the second set of temporal pixels comprises a set of pixel values at an associated position from each image of the second set of images;
• determining one or more derived values based on values of the temporal pixels in the first temporal pixel image, the second temporal pixel image, or both;
• determining, based on the first temporal pixel image and the second temporal pixel image, a correspondence density indicative of a relationship between a number of a set of identified correspondences between image points of the first set of images and image points of the second set of images and a number of possible correspondences;
• determining, based on the one or more derived values and the correspondence density, an indication of whether there is a likelihood of motion of an object in the scene;
• and
• based on determining that there is a likelihood of motion of the object in the scene, determine a position of the object in the scene at a future time subsequent to a time related to when the first set of images and the second set of images were captured.