Systems and methods for image feature extraction
AssigneeTexas Instruments Incorporated
InventorsDeepak Kumar Poddar, Soyeb Nagori, Hrushikesh Tukaram Garud, Kumar Desappan
Life of this patent
Estimating the term…
Abstract
An example image feature extraction system comprises an encoder neural network having a first set of layers and a decoder neural network having a second set of layers and a third set of layers. The encoder neural network receives an input image, processes the input image through the first set of layers, and computes an encoded feature map based on the input image. The decoder neural network receives the encoded feature map, processes the encoded feature map through the second set of layers to compute a keypoint score map, and processes the encoded feature map through at least a portion of the third set of layers to compute a feature description map.
Claims17
A system comprising:
- an encoder neural network having a first set of layers, the encoder neural network configured to:
- receive an input image,
- process the input image through the first set of layers, and
- compute an encoded feature map based on the input image;
- and
- a decoder neural network coupled to the encoder neural network, the decoder neural network having a second set of layers and a third set of layers and configured to:
- receive the encoded feature map,
- process the encoded feature map through the second set of layers to compute a keypoint score map, and
- process the encoded feature map through at least a portion of the third set of layers to compute a feature description map.
One or more non-transitory machine-readable media having machine-readable instructions executable by a processor, the machine-readable instructions comprising instructions for:
- processing an input image through a first set of layers of an encoder neural network;
- computing an encoded feature map based on the input image, using the first set of layers;
- processing the encoded feature map through a second set of layers of a decoder neural network to compute a keypoint score map;
- and
- processing the encoded feature map through at least a portion of a third set of layers of the decoder neural network to compute a feature description map.
Dependencies are inferred from the claim text.