Bounding box generation for object detection
AssigneeGM Cruise Holdings LLC
InventorsMatthias Wisniowski, Radu Dondera, Bin Ma, Siddharth Raina
Life of this patent
Estimating the term…
Abstract
The subject disclosure relates to techniques for inserting of bounding boxes around image objects. A process of the disclosed technology can include steps for receiving, from a first data set, an image comprising a first image object, processing the image to identify a pixel region associated with the first image object, and placing a first bounding box around the first image object based on the identified pixel region. In some aspects, the process further includes steps for receiving a user input comprising an indication of whether the first bounding box is accurately placed around the first image object. Systems and machine-readable media are also provided.
Claims18
A computer-implemented method for performing image-object detection, comprising:
- receiving, from a first data set, an image comprising a first image object;
- processing the image to identify a pixel region associated with the first image object;
- placing a first bounding box around the first image object based on the identified pixel region;
- receiving a user input comprising an indication of whether the first bounding box is accurately placed around the first image object, wherein the user input indicates a centroid location of the pixel region associated with the first image object;
- automatically resizing the first bounding box based on the user input;
- and
- automatically identifying one or more semantic labels that are associated with the first image object.
A system for performing image-object detection comprising:
- one or more processors;
- and
- a computer-readable medium comprising instructions stored therein, which when executed by the processors, cause the processors to perform operations comprising:
- receiving, from a first data set, an image comprising a first image object;
- processing the image to identify a pixel region associated with the first image object;
- placing a first bounding box around the first image object based on the identified pixel region;
- receiving a user input comprising an indication of whether the first bounding box is accurately placed around the first image object, wherein the user input indicates a centroid location of the pixel region associated with the first image object;
- automatically resizing the first bounding box based on the user input;
- and
- automatically identifying one or more semantic labels that are associated with the first image object.
A non-transitory computer-readable storage medium comprising instructions stored therein, which when executed by one or more processors, cause the processors to perform operations comprising:
- receiving, from a first data set, an image comprising a first image object;
- processing the image to identify a pixel region associated with the first image object;
- placing a first bounding box around the first image object based on the identified pixel region;
- receiving a user input comprising an indication of whether the first bounding box is accurately placed around the first image object, wherein the user input indicates a centroid location of the pixel region associated with the first image object;
- automatically resizing the first bounding box based on the user input;
- and
- automatically identifying one or more semantic labels that are associated with the image object.
Dependencies are inferred from the claim text.