JPEG classifier
A method, device and computer program product, the method comprising: obtaining access to a classifier trained upon a multiplicity of sets of decoded coefficients; obtaining a set of block coefficients associated with at least a part of the compressed image; and applying the classifier to the set of block coefficients, to obtain a classification of the compressed image.
1. A method for classifying an image, comprising:
obtaining access to a classifier trained upon a multiplicity of sets of decoded coefficients;
obtaining, by partially decompressing a compressed version of the image, one or more sets of first block coefficients associated with one or more first image blocks of a plurality of image blocks, each image block of the plurality of image blocks representing a respective region of the image, wherein each set of first block coefficients of the one or more sets of first block coefficients is associated with a respective first image block of the one or more first image blocks;
applying the classifier to the one or more sets of first block coefficients to obtain a classification of the compressed version of the image and a probability of the compressed version of the image meeting the classification; and
selectively obtaining, from the compressed version of the image, one or more sets of second block coefficients associated with one or more second image blocks of the plurality of image blocks based at least in part on the probability of the compressed version of the image meeting the classification.
2. The method of claim 1 , wherein the one or more sets of first block coefficients is received from a capture device configured for:
capturing the image;
dividing the image into the plurality of image blocks;
compressing the image to determine a respective set of block coefficients for each image block of the plurality of image blocks; and
outputting the set of block coefficients for at least one first image block included in the one or more first image blocks no later than the image is fully compressed, wherein the one or more first image blocks are a subset of the plurality of image blocks.
3. The method of claim 1 , wherein partially decompressing the compressed version of the image and applying the classifier to the one or more sets of first block coefficients takes at most 50% of processing steps required for full decompression and classification of the compressed version of the image.
4. The method of claim 1 , wherein partially decompressing the compressed version of the image and applying the classifier to the one or more sets of first block coefficients takes at most 10% of processing steps required for full decompression and classification of the compressed version of the image.
5. The method of claim 1 , wherein partially decompressing the compressed version of the image and applying the classifier to the one or more sets of first block coefficients is performed in hardware.
6. The method of claim 1 , wherein the method is used for security or privacy protection purposes.
7. The method of claim 1 , wherein the compressed version of the image is a Joint Photographic Experts Group (JPEG) image.
8. A device comprising:
an image capture device for capturing an image;
a memory device for storing a compressed image being the image as compressed; and
a processor, executing:
obtaining access to a classifier trained upon a multiplicity of sets of decoded coefficients;
obtaining, by partially decompressing the compressed image, one or more sets of first block coefficients associated with one or more first image blocks of a plurality of image blocks, each image block of the plurality of image blocks representing a respective region of the image, wherein each set of first block coefficients of the one or more sets of first block coefficients is associated with a respective first image block of the one or more first image blocks;
applying the classifier to the one or more sets of first block coefficients to obtain a classification of the compressed image and a probability of the compressed image meeting the classification; and
selectively obtaining, from the compressed image, one or more sets of second block coefficients associated with one or more second image blocks of the plurality of image blocks based at least in part on the probability of the compressed image meeting the classification.
9. The device of claim 8 , wherein the one or more sets of first block coefficients is received from the image capture device, wherein the image capture device is configured for:
capturing the image;
dividing the image into the plurality of image blocks;
compressing the image to determine a respective set of block coefficients for each image block of the plurality of image blocks; and
outputting the set of block coefficients for at least one first image block included in the one or more first image blocks no later than the image is fully compressed, wherein the one or more first image blocks are a subset of the plurality of image blocks.
10. The device of claim 8 , wherein partially decompressing the compressed image and applying the classifier to the one or more sets of first block coefficients takes at most 10% of the processing steps required for full decompression and classification of the compressed image.
11. The device of claim 8 , wherein the compressed image is a Joint Photographic Experts Group (JPEG) image.
12. A computer program product comprising a non-transitory computer readable storage medium retaining program instructions, which program instructions when read by a processor, cause the processor to perform a method comprising:
obtaining access to a classifier trained upon a multiplicity of sets of decoded coefficients;
obtaining, by partially decompressing a compressed version of an image, one or more sets of first block coefficients associated with one or more first image blocks of a plurality of image blocks, each image block of the plurality of image blocks representing a respective region of the image, wherein each set of first block coefficients of the one or more sets of first block coefficients is associated with a respective first image block of the one or more first image blocks; and
applying the classifier to the one or more sets of first block coefficients to obtain a classification of the compressed version of the image and a probability of the compressed version of the image meeting the classification; and
selectively obtaining, from the compressed version of the image, one or more sets of second block coefficients associated with one or more second image blocks of the plurality of image blocks based at least in part on the probability of the compressed version of the image meeting the classification.