User interest based packetization
An image coding mechanism is disclosed. The mechanism includes determining a region of interest in an image. The image is partitioned into a plurality of blocks. A transform is applied to the blocks to obtain a plurality of coefficients. The coefficients are assigned into quality layers of decreasing priority. Each subsequent layer includes data to incrementally increase quality of a reconstructed image. Region of interest coefficients are positioned in one or more image data packets by quality layer in order of decreasing priority. Remaining coefficients are positioned in the image data packets with a lower priority than the region of interest coefficients. The remaining coefficients are positioned by quality layer in order of decreasing priority. The image data packets are stored and transmitted over a network.
1. A method implemented by a network device, the method comprising:
determining, by at least one processor of an encoder, a region of interest in an image;
partitioning, by the at least one processor, the image into a plurality of blocks;
applying, by the at least one processor, a transform to the blocks to obtain a plurality of coefficients;
assigning, by the at least one processor, the coefficients into quality layers of decreasing priority, each subsequent layer including data to incrementally increase quality of a reconstructed image;
positioning, by the at least one processor, region of interest coefficients in one or more image data packets by quality layer and in an order of decreasing priority;
positioning, by the at least one processor, remaining coefficients in the image data packets with a lower priority than the region of interest coefficients, the remaining coefficients positioned by quality layer and in the order of decreasing priority;
storing, by a memory of the encoder, the image data packets; and
transmitting, by a transmitter, the image data packets over a network.
2. The method of claim 1 , wherein the region of interest coefficients comprise coefficients in a most significant quality layer of a region of interest and coefficients in a less significant quality layer of the region of interest, wherein the remaining coefficients comprise coefficients in a most significant quality layer of a remaining region and coefficients in a less significant quality layer of the remaining region, wherein the coefficients in the most significant quality layer of the region of interest are positioned in the image data packets with a higher priority than the coefficients in the most significant quality layer of the remaining region, wherein the coefficients in the most significant quality layer of the remaining region are positioned in the image data packets with a higher priority than the coefficients in the less significant quality layer of the region of interest, and wherein the coefficients in the less significant quality layer of the region of interest are positioned in the image data packets with a higher priority than the coefficients in the less significant quality layer of the remaining region.
3. The method of claim 1 , wherein the region of interest coefficients are positioned in the image data packets with a higher priority than the remaining coefficients.
4. The method of claim 1 , further comprising encoding in a header of the image data packets, by the at least one processor, a location of the region of interest coefficients in the reconstructed image.
5. The method of claim 1 , further comprising encoding in a header of the image data packets, by the at least one processor, an indication of a number of blocks in the region of interest.
6. The method of claim 1 , further comprising encoding in a header of the image data packets, by the at least one processor, an indication of a number of blocks outside the region of interest.
7. The method of claim 6 , further comprising receiving, by a receiver, a loss tolerance for the region of interest from a decoder, wherein the region of interest coefficients are assigned into quality layers and positioned in the image data packets based on the loss tolerance for the region of interest.
8. The method of claim 7 , further comprising encoding in a header of the image data packets, by the at least one processor, a loss tolerance outside the region of interest.
9. The method of claim 8 , further comprising receiving, by the receiver, the loss tolerance outside the region of interest from the decoder, wherein the remaining coefficients are organized into quality layers and positioned in the image data packets based on the loss tolerance outside the region of interest.