Back to results
US 2022/0277159 A1Published application
Object Detection Network with Spatial Uncertainty
AssigneeArm Limited
InventorsPartha Prasun Maji, Tiago Manuel Lourenco Azevedo
Life of this patent
Estimating the term…
Abstract
A hardware accelerator for an object detection network and a method for detecting an object are provided. The present disclosure provides robust object detection that advantageously augments traditional deterministic bounding box predictions with spatial uncertainties for various computer vision applications, such as, for example, autonomous driving, robotic surgery, etc.
Claims20
A hardware accelerator for object detection, comprising:
- a backbone feature extraction (BFE) module configured to:
- receive image data including at least one object, and
- generate a number of feature tensors based on the image data, each feature tensor having a different scale;
- a feature detection (FD) module configured to:
- receive the feature tensors from the BFE module,
- perform a number of detection passes, each detection pass including:
- apply a random dropout mask to the feature tensors, and
- for each feature tensor, determine scale-dependent bounding box spatial data including mean corner co-ordinates and covariance matrices for the corner co-ordinates for a number of scale-dependent bounding boxes for the object, and
- generate an intermediate tensor associated with each feature tensor, each intermediate tensor including the scale-dependent bounding box spatial data for each detection pass for the associated feature tensor;
- and
- a filter and averaging (FA) module configured to:
- receive the intermediate tensors from the FD module,
- determine scale-independent bounding box spatial data based on the intermediate tensors, the scale-independent bounding box spatial data including mean corner co-ordinates and covariance matrices for the corner co-ordinates for a scale-independent bounding box for the object, and
- generate an output tensor including the scale-independent bounding box spatial data.
11 more elements
A method for detecting an object, comprising:
- at a backbone feature extraction (BFE) module:
- receiving image data including at least one object;
- generating a number of feature tensors based on the image data, each feature tensor having a different scale;
- at a feature detection (FD) module:
- receiving the feature tensors from the BFE module;
- performing a number of detection passes, each detection pass including:
- applying a random dropout mask to the feature tensors;
- for each feature tensor, determining scale-dependent bounding box spatial data including mean corner co-ordinates and covariance matrices for the corner co-ordinates for a number of scale-dependent bounding boxes for the object;
- generating an intermediate tensor associated with each feature tensor, each intermediate tensor including the scale-dependent bounding box spatial data for each detection pass for the associated feature tensor;
- at a filter and averaging (FA) module:
- receiving the intermediate tensors from the FD module;
- determining scale-independent bounding box spatial data based on the intermediate tensors, the scale-independent bounding box spatial data including mean corner co-ordinates and covariance matrices for the corner co-ordinates for a scale-independent bounding box for the object;
- and
- generating an output tensor including the scale-independent bounding box spatial data.
11 more elements
Dependencies are inferred from the claim text.