Abstract: Embodiments are generally directed to methods and apparatuses for dynamic normalization and relay in a neural network. An embodiment of an apparatus for dynamic normalization and relay in a neural network including a hyper normalization layer comprises: a compute engine to: generate a hidden state and a cell state for the hyper normalization layer based on an input feature map for the hyper normalization layer as well as a previous hidden state and a previous cell state; and normalize the input feature map in the hyper normalization layer with the hidden state and the cell state for the hyper normalization layer.

Abstract: The present disclosure provides an apparatus and method of guided neural network model for image processing. An apparatus may comprise a guidance map generator, a synthesis network and an accelerator. The guidance map generator may receive a first image as a content image and a second image as a style image, and generate a first plurality of guidance maps and a second plurality of guidance maps, respectively from the first image and the second image. The synthesis network may synthesize the first plurality of guidance maps and the second plurality of guidance maps to determine guidance information. The accelerator may generate an output image by applying the style of the second image to the first image based on the guidance information.

Abstract: Techniques and apparatus for generating dense natural language descriptions for video content are described. In one embodiment, for example, an apparatus may include at least one memory and logic, at least a portion of the logic comprised in hardware coupled to the at least one memory, the logic to receive a source video comprising a plurality of frames, determine a plurality of regions for each of the plurality of frames, generate at least one region-sequence connecting the determined plurality of regions, apply a language model to the at least one region-sequence to generate description information comprising a description of at least a portion of content of the source video. Other embodiments are described and claimed.

Abstract: Systems, methods, apparatuses, and computer program products to receive a plurality of binary weight values for a binary neural network sampled from a policy neural network comprising a posterior distribution conditioned on a theta value. An error of a forward propagation of the binary neural network may be determined based on a training data and the received plurality of binary weight values. A respective gradient value may be computed for the plurality of binary weight values based on a backward propagation of the binary neural network. The theta value for the posterior distribution may be updated using reward values computed based on the gradient values, the plurality of binary weight values, and a scaling factor.

Abstract: Methods and systems for advanced and augmented training of deep neural networks (DNNs) using synthetic data and innovative generative networks. A method includes training a DNN using synthetic data, training a plurality of DNNs using context data, associating features of the DNNs trained using context data with features of the DNN trained with synthetic data, and generating an augmented DNN using the associated features.

Abstract: Embodiments are generally directed to sparse 3D convolution acceleration in a convolutional layer of an artificial neural network model. An embodiment of an apparatus includes one or more processors including a graphics processor to process data; and a memory for storage of data, including feature maps. The one or more processors are to provide for sparse 3D convolution acceleration by applying a shared 3D convolutional kernel/filter to an input feature map to produce an output feature map, including increasing sparsity of the input feature map by partitioning it into multiple disjoint input groups; generation of multiple disjoint output groups corresponding to the input groups by performing a convolution calculation represented by the shared 3D convolutional kernel/filter on all feature values associated with active/valid voxels of each input group to produce corresponding feature values within corresponding output groups; and outputting the output feature map by sequentially stacking the output groups.

Abstract: Apparatuses, methods, and GPUs are disclosed for universal loss-error-aware quantization (ULQ) of a neural network (NN). In one example, an apparatus includes data storage to store data including activation sets and weight sets, and a network processor coupled to the data storage. The network processor is configured to implement the ULQ by constraining a low-precision NN model based on a full-precision NN model, to perform a loss-error-aware activation quantization to quantize activation sets into ultra-low-bit versions with given bit-width values, to optimize the NN with respect to a loss function that is based on the full-precision NN model, and to perform a loss-error-aware weight quantization to quantize weight sets into ultra-low-bit versions.

Abstract: Techniques related to capturing 3D faces using image and temporal tracking neural networks and modifying output video using the captured 3D faces are discussed. Such techniques include applying a first neural network to an input vector corresponding to a first video image having a representation of a human face to generate a morphable model parameter vector, applying a second neural network to an input vector corresponding to a first and second temporally subsequent to generate a morphable model parameter delta vector, generating a 3D face model of the human face using the morphable model parameter vector and the morphable model parameter delta vector, and generating output video using the 3D face model.

Abstract: An example apparatus for mining multi-scale hard examples includes a convolutional neural network to receive a mini-batch of sample candidates and generate basic feature maps. The apparatus also includes a feature extractor and combiner to generate concatenated feature maps based on the basic feature maps and extract the concatenated feature maps for each of a plurality of received candidate boxes. The apparatus further includes a sample scorer and miner to score the candidate samples with multi-task loss scores and select candidate samples with multi-task loss scores exceeding a threshold score.

Abstract: Methods and systems for budgeted and simplified training of deep neural networks (DNNs) are disclosed. In one example, a trainer is to train a DNN using a plurality of training sub-images derived from a down-sampled training image. A tester is to test the trained DNN using a plurality of testing sub-images derived from a down-sampled testing image. In another example, in a recurrent deep Q-network (RDQN) having a local attention mechanism located between a convolutional neural network (CNN) and a long-short time memory (LSTM), a plurality of feature maps are generated by the CNN from an input image. Hard-attention is applied by the local attention mechanism to the generated plurality of feature maps by selecting a subset of the generated feature maps. Soft attention is applied by the local attention mechanism to the selected subset of generated feature maps by providing weights to the selected subset of generated feature maps in obtaining weighted feature maps.

Abstract: Techniques and apparatus for generating dense natural language descriptions for video content are described. In one embodiment, for example, an apparatus may include at least one memory and logic, at least a portion of the logic comprised in hardware coupled to the at least one memory, the logic to receive a source video comprising a plurality of frames, determine a plurality of regions for each of the plurality of frames, generate at least one region-sequence connecting the determined plurality of regions, apply a language model to the at least one region-sequence to generate description information comprising a description of at least a portion of content of the source video. Other embodiments are described and claimed.

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.

Abstract: Region proposal is described for image regions that include objects of interest. Feature maps from multiple layers of a convolutional neural network model are used. In one example a digital image is received and buffered. Layers of convolution are performed on the image to generate feature maps. The feature maps are reshaped to a single size. The reshaped feature maps are grouped by sequential concatenation to form a combined feature map. Region proposals are generated using the combined feature map by scoring bounding box regions of the image. Objects are detected and classified objects in the proposed regions using the feature maps.

Abstract: A convolutional neural network framework is described that uses reverse connection and obviousness priors for object detection. A method includes performing a plurality of layers of convolutions and reverse connections on a received image to generate a plurality of feature maps, determining an objectness confidence for candidate bounding boxes based on outputs of an objectness prior, determining a joint loss function for each candidate bounding box by combining an objectness loss, a bounding box regression loss and a classification loss, calculating network gradients over positive boxes and negative boxes, updating network parameters within candidate bounding boxes using the joint loss function, repeating performing the convolutions through to updating network parameters until the training converges, and outputting network parameters for object detection based on the training images.

Abstract: Techniques are provided for neural network based, human attribute recognition, guided by anatomical key-points and statistic correlation models. Attributes include characteristics that can be visibly identified or inferred from an image, such as gender, hairstyle, clothing style, etc. A methodology implementing the techniques according to an embodiment includes applying an attribute feature extraction (AFE) convolutional neural network (CNN) to an image of a human to generate attribute feature maps based on the image. The method further includes applying a key-point guided proposal (KPG) CNN to the image of the human to generate proposed hierarchical regions of the image based on associated anatomical key-points.

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 comprising a plurality of gated dense pyramid (GDP) blocks to be trained to generate semantic labels for each pixel in the received image. The apparatus further includes a generator to generate a segmented image based on the generated semantic labels.

Abstract: An apparatus for dynamic emotion recognition in unconstrained scenarios is described herein. The apparatus comprises a controller to pre-process image data and a phase-convolution mechanism to build lower levels of a CNN such that the filters form pairs in phase. The apparatus also comprises a phase-residual mechanism configured to build middle layers of the CNN via plurality of residual functions and an inception-residual mechanism to build top layers of the CNN by introducing multi-scale feature extraction. Further, the apparatus comprises a fully connected mechanism to classify extracted features.

Abstract: Combinatorial shape regression is described as a technique for face alignment and facial landmark detection in images. As described stages of regression may be built for multiple ferns for a facial landmark detection system. In one example a regression is performed on a training set of images using face shapes, using facial component groups, and using individual face point pairs to learn shape increments for each respective image in the set of images. A fern is built based on this regression. Additional regressions are performed for building additional ferns. The ferns are then combined to build the facial landmark detection system.

Abstract: An example apparatus for mining multi-scale hard examples includes a convolutional neural network to receive a mini-batch of sample candidates and generate basic feature maps. The apparatus also includes a feature extractor and combiner to generate concatenated feature maps based on the basic feature maps and extract the concatenated feature maps for each of a plurality of received candidate boxes. The apparatus further includes a sample scorer and miner to score the candidate samples with multi-task loss scores and select candidate samples with multi-task loss scores exceeding a threshold score.

Abstract: Methods and apparatus for multi-task recognition using neural networks are disclosed. An example apparatus includes a filter engine to generate a facial identifier feature map based on image data, the facial identifier feature map to identify a face within the image data. The example apparatus also includes a sibling semantic engine to process the facial identifier feature map to generate an attribute feature map associated with a facial attribute. The example apparatus also includes a task loss engine to calculate a probability factor for the attribute, the probability factor identifying the facial attribute. The example apparatus also includes a report generator to generate a report indicative of a classification of the facial attribute.