Semantic image segmentation using gated dense pyramid blocks
AssigneeIntel Corporation
InventorsLibin Wang, Anbang Yao, Jianguo Li, Yurong Chen
Life of this patent
Estimating the term…
Abstract
An example apparatus for semantic image segmentation includes a receiver to receive an image to be segmented. The apparatus also includes a gated dense pyramid network including a plurality of gated dense pyramid (GDP) blocks to be trained to generate semantic labels for respective pixels in the received image. The apparatus further includes a generator to generate a segmented image based on the generated semantic labels.
Claims25
At least one computer readable memory or storage device comprising computer readable instructions that, when executed, cause at least one processor to at least:
- process an input image with a trained, gated dense pyramid (GDP) network to generate semantic labels for respective pixels in the input image, the GDP network including a plurality of GDP blocks, a first one of the GDP blocks to include at least one of (i) a plurality of middle layers coupled by dense connections including a forward dense connection and a backward dense connection, (ii) a forward dense connection with a middle layer having a decreased scale to generate down-sampled feature maps, (iii) a global contextual feature to smooth prediction results, or (iv) a backward dense connection with a middle layer to generate up-sampled feature maps;
- and
- generate, based on the generated semantic labels, a segmented image corresponding to the input image.
At least one computer readable memory or storage device comprising computer readable instructions that, when executed, cause at least one processor to at least:
- process an input image with a convolutional neural network to form a pre-processed image;
- process the pre-processed image with a trained, gated dense pyramid (GDP) network to generate semantic labels for respective pixels in the pre-processed image, the GDP network including a plurality of GDP blocks;
- and
- generate, based on the generated semantic labels, a segmented image corresponding to the input image.
At least one computer readable memory or storage device comprising computer readable instructions that, when executed, cause at least one processor to at least:
- forward train a gated dense pyramid (GDP) block of a gated dense pyramid network;
- backward train the GDP block of the gated dense pyramid network with a gated function;
- process an input image with the GDP network to generate semantic labels for respective pixels in the input image;
- and
- generate a segmented image based on the semantic labels generated for the respective pixels in the input image.
An apparatus for semantic image segmentation, the apparatus comprising:
- at least one memory;
- instructions in the apparatus;
- and
- at least one processor to execute the instructions to at least:
- forward train a gated dense pyramid (GDP) block of a gated dense pyramid network;
- backward train the GDP block of the gated dense pyramid network with a gated function;
- process an input image with the GDP network to generate semantic labels for respective pixels in the input image;
- and
- generate a segmented image based on the semantic labels generated for the respective pixels in the input image.
Dependencies are inferred from the claim text.