Obfuscating portions of video data

A computer-implemented method, comprising:

• receiving, from a first device, a first depth data that includes depth values corresponding to an operator;
• receiving, from a second device, a first video frame that includes a plurality of pixel data;
• generating a three-dimensional model (“3D model”) based at least in part on the first depth data, wherein the 3D model represents a plurality of points in three-dimensional space corresponding to at least the operator;
• align the first video frame with points of the 3D model based at least in part on the plurality of points in the three-dimensional space;
• determining, for a first plurality of pixels of the plurality of pixels of the first video frame and based at least in part on the alignment of the first video frame with points of the 3D model, that the first plurality of pixels corresponds to the operator;
• generating an obfuscation video frame that includes obfuscation pixels corresponding to the first plurality of pixels;
• and
• concurrently presenting the first video frame and the obfuscation video frame such that the obfuscation video frame is overlaid atop the first video frame and the obfuscation pixels obfuscate each of the first plurality of pixels.

An apparatus, comprising:

• a shelf;
• a first device aligned to include at least a portion of the shelf within a first field of view;
• a second device aligned to include at least a portion of the shelf within a second field of view, wherein at least a portion of the second field of view at least partially overlaps at least a portion of the first field of view;
• and
• a computing device in communication with at least the first device and the second device,
• wherein the computing device is configured to at least:
• receive, from the first device, depth data that includes depth values corresponding to an operator positioned proximate the shelf;
• receive, from the second device, a first video frame that includes a plurality of pixels;
• as the first video frame is received:
• generate a three-dimensional model (“3D model”) based at least in part on the depth data, wherein the 3D model represents a plurality of points in three-dimensional space corresponding to at least the operator;
• align the first video frame with points of the 3D model based at least in part on the plurality of points in the three-dimensional space;
• determine, for a first plurality of pixels of the plurality of pixels and based at least in part on the alignment of the first video frame with points of the 3D model, that the first plurality of pixels correspond to the operator;
• generate an obfuscation video frame that includes obfuscation pixels corresponding to the first plurality of pixels;
• and
• concurrently present the first video frame and the obfuscation video frame such that the obfuscation video frame is overlaid atop the first video frame and the obfuscation pixels obfuscate each of the first plurality of pixels.