Neural image compression with controllable spatial bit allocation
AssigneeQUALCOMM Incorporated
InventorsYang YANG, Hoang Cong Minh LE, Yinhao ZHU, Reza POURREZA, Amir SAID, Yizhe ZHANG, Taco Sebastiaan COHEN
Life of this patent
Estimating the term…
Abstract
A processor-implemented method for image compression using an artificial neural network (ANN) includes receiving, at an encoder of the ANN, an image and a spatial segmentation map corresponding to the image. The spatial segmentation map indicates one or more regions of interest. The encoder compresses the image according to a controllable spatial bit allocation. The controllable spatial bit allocation is based on a learned quantization bin size.
Claims20
A processor-implemented method, comprising:
- receiving, at an encoder of an artificial neural network, an image and a spatial segmentation map corresponding to the image, the spatial segmentation map indicating one or more regions of interest;
- and
- compressing, via the encoder, the image according to a controllable spatial bit allocation, the controllable spatial bit allocation is based on a learned quantization bin size.
A processor-implemented method, comprising:
- receiving, at a decoder of an artificial neural network, a bit stream from an encoder;
- decoding, via the decoder, the bit stream to recover a quantized latent representation of an image corresponding to a spatial segmentation map, the spatial segmentation map indicating one or more regions of interest;
- applying reciprocal gain control elements to the quantized latent representation to produce the spatial segmentation map, the reciprocal gain control elements having a gain value corresponding to a learned quantization bin size;
- and
- generating, via the decoder, a reconstructed image based on the spatial segmentation map.
An apparatus, comprising:
- a memory;
- and
- at least one processor coupled to the memory, the at least one processor configured:
- to receive, at an encoder of an artificial neural network, an image and a spatial segmentation map corresponding to the image, the spatial segmentation map indicating one or more regions of interest;
- and
- to compress, via the encoder, the image according to a controllable spatial bit allocation, the controllable spatial bit allocation is based on a learned quantization bin size.
An apparatus, comprising:
- a memory;
- and
- at least one processor coupled to the memory, the at least one processor configured:
- to receive, at a decoder of an artificial neural network, a bit stream from an encoder;
- to decode, via the decoder, the bit stream to recover a quantized latent representation of an image corresponding to a spatial segmentation map, the spatial segmentation map indicating one or more regions of interest;
- to apply reciprocal gain control elements to the quantized latent representation to produce the spatial segmentation map, the reciprocal gain control elements having a gain value corresponding to a learned quantization bin size;
- and
- to generate, via the decoder, a reconstructed image based on the spatial segmentation map.
Dependencies are inferred from the claim text.