Abstract: A method, apparatus, and a non-transitory computer-readable storage medium for image denoising. The method may include obtaining a raw image captured by a camera. The method may also include obtaining a color modeled image based on the raw image. The method may further include obtaining a subsampled raw image based on the raw image. The method may also include obtaining a denoised image based on a neural network processing the color modeled image and the subsampled raw image.

Abstract: A method, apparatus, and a non-transitory computer-readable storage medium for generating heterogenous platform code. The method may obtain a neural network model. The neural network model may be programed to run on at least one platform. The method may also obtain an initial intermediate representation (IR) code by encoding the neural network model, and obtain a target IR code by adding decorations to the initial IR code based on a target platform. The method may also output an executable code optimized to run on the target platform by decoding the target IR code.

Abstract: Systems and methods are provided for automatic speech recognition. In the method, the system obtains a padded sequence by processing a plurality of acoustic signals. The system compresses the padded sequence by reducing the size of the padded sequence to obtain a compressed sequence. The system inputs the compressed sequence into a pre-trained encoder neural network to obtain an encoded sequence and then decompresses the encoded sequence by recovering the encoded sequence to an original sequential ordering. The system inputs the encoded sequence to a decoding module to obtain recognition texts.

Abstract: An automatic speech recognition system and a method thereof are provided. The system includes an encoder and a decoder. The encoder comprises a plurality of encoder layers. At least one encoder layer includes a plurality of encoder sublayers fused into one or more encoder kernels. The system further comprises a first pair of ping-pong buffers communicating with the one or more encoder kernels. The decoder comprises a plurality of decoder layers. At least one decoder layer includes a plurality of decoder sublayers fused into one or more decoder kernels. The decoder receives a decoder output related to the encoder output and generates a decoder output. The encoder sends the decoder output to a beam search kernel.

Abstract: Methods and apparatuses are provided for movie and television series video data analysis. The method includes: gathering and reading, by a processor, a plurality of input movies; removing a video border of each input movie; splitting the input movie into short clips, based on accuracy and efficiency requirements of different analyzing models; assessing attributes of each input movie by analyzing, with the different analyzing models, the input movie, the short clips cut from the input movie, and the frame images extracted from the input movie; and summarizing the plurality of input movies based on matching and integrating the attributes assessed for each input movie.

Abstract: A method for training a neural network system for generating paired low resolution (LR) and high resolution (HR) images, the neural network system, an apparatus, and a non-transitory computer-readable storage medium thereof are provided. The method includes that a first generator in the neural network system generates a LR image based on a random vector; a second generator in the neural network system generates a HR image based on the random vector, where the HR image is paired with the LR image; obtaining a plurality of losses based on the LR image and the HR image; and updating the first generator based on the plurality of losses.

Abstract: A neural network system for restoring images, a method and a non-transitory computer-readable storage medium thereof are provided. The neural network system includes an encoder and a generative adversarial network (GAN) prior network. The encoder includes a plurality of encoder blocks, where each encoder block includes at least one transformer block and one convolution layer, where the encoder receives an input image and generates a plurality of encoder features and a plurality of latent vectors. Additionally, the GAN prior network includes a plurality of pre-trained generative prior layers, where the GAN prior network receives the plurality of encoder features and the plurality of latent vectors from the encoder and generates an output image with super-resolution.

Abstract: Methods and apparatuses are provided for movie and television series video data analysis. The method includes: gathering and reading, by a processor, a plurality of input movies; removing a video border of each input movie; splitting the input movie into short clips, based on accuracy and efficiency requirements of different analyzing models; assessing attributes of each input movie by analyzing, with the different analyzing models, the input movie, the short clips cut from the input movie, and the frame images extracted from the input movie; and summarizing the plurality of input movies based on matching and integrating the attributes assessed for each input movie.

Abstract: A method and an apparatus for length-aware local tiling in a sparse attention module in a transformer in heterogeneous devices are provided. The method includes that a heterogeneous device including one or more GPUs: divides a transformed sparsity mask into a plurality of first tiles and obtaining one or more effective first tiles from the plurality of first tiles, where each effective first tile includes at least one non-zero element; loads the one or more effective first tiles into a shared memory in the one or more GPUs and loads a plurality of elements in a first matrix corresponding to the one or more effective first tiles into the shared memory; and performs multiplication by a first sampled dense-dense matrix multiplication (SDDMM) kernel in the sparse attention module in the transformer by fetching the one or more effective first tiles and the plurality of elements from the shared memory.

Abstract: A method, apparatus, and a non-transitory computer-readable storage medium for generating heterogenous platform code. The method may obtain a neural network model. The neural network model may be programed to run on at least one platform. The method may also obtain an initial intermediate representation (IR) code by encoding the neural network model, and obtain a target IR code by adding decorations to the initial IR code based on a target platform. The method may also output an executable code optimized to run on the target platform by decoding the target IR code.

Abstract: A method and an apparatus for accelerating a transformer with a sparse attention pattern are provided. The method includes that a heterogeneous device including one or more GPUs loads a first matrix, a second matrix, and a transformed sparsity mask into a first sampled dense-dense matrix multiplication (SDDMM) kernel in a sparse attention module in the transformer and generates a first output based on the first matrix, the second matrix, and the transformed sparsity mask by the first SDDMM kernel, generates a second output by a softmax kernel in the sparse attention module based on the first output, loads the second output, a third matrix, and the transformed sparsity mask into a matrix multiplication kernel in the sparse attention module, and generates an output of the sparse attention module.

Abstract: A method, apparatus, and a non-transitory computer-readable storage medium for image denoising. The method may include obtaining a raw image captured by a camera. The method may also include obtaining a color modeled image based on the raw image. The method may further include obtaining a subsampled raw image based on the raw image. The method may also include obtaining a denoised image based on a neural network processing the color modeled image and the subsampled raw image.

Abstract: A method and an apparatus for video processing are provided. The method includes that a decoding terminal receives a plurality of coded video frames coded using one or more generative adversarial networks (GANs), receives network parameters related to the one or more GANs, and decodes the plurality of coded video frames using GANs based on the network parameters. Further, the one or more GANs respectively implement one or more video coding functions including reference-frame coding, motion-compensated frame prediction, and residue-frame coding.

Abstract: Methods and apparatuses are provided for editing and generating a short video based upon a long video. The method includes: obtaining a plurality of short source video clips as candidate video clips; obtaining attributes of each short source video clip; obtaining a plurality of target base videos according to a target genre, processing the plurality of target base videos by splitting each target base video into a plurality of short target base video clips, and obtaining attributes of each short target base video clip; selecting short target video clips from the plurality of short source video clips, based on distribution of the attributes obtained for the plurality of the short source video clips and the plurality of short target base video clips; and editing and combining the short target video clips selected from the plurality of short source video clips, to obtain a target video.

Abstract: A method, apparatus, and a non-transitory computer-readable storage medium for sub-band image reconstruction. The method may include obtaining an image captured by a camera. The method may also obtain a transform image based on the image captured by the camera. The transform image may be in a transform domain. The method may further obtain decomposed image components of the transform image. The decomposed image components may include a low frequency component and at least one high frequency component.

Abstract: A neural network training method, an image processing method, and apparatuses thereof are provided. The neural network training method includes obtaining a first domain image and a second domain image, where the first domain image and the second domain image are unpaired images in different domains; obtaining a scaled first domain image by scaling, at an iteration, the first domain image; obtaining a training patch by cropping the scaled first domain image, where each training patch has a same number of pixels with different contents; inputting the training patch into the neural network at the iteration, and outputting an output patch; calculating a contrastive loss based on a query sub-patch and negative sub-patches selected from the training patch and a corresponding positive sub-patch selected from the output patch; and updating model parameters of the neural network based on the contrastive loss and a generative adversarial network loss.

Abstract: Methods and apparatuses are provided for temporal profiling for neural network quantization. The method includes: obtaining a neural network that comprises anode connected to different paths at different time periods; obtaining node outputs for the node at the different time periods; determining statistic properties of the node outputs at the different time periods; and determining activation ranges of the node outputs based on the statistic properties.

Abstract: Methods and devices are provided for implementing efficient general deconvolution Implementation on hardware accelerator.

Abstract: A method and an apparatus for training a transferable vision transformer (TVT) for unsupervised domain adaption (UDA) in heterogeneous devices are provided. The method includes that a heterogeneous device including one or more graphic processing units (GPUs) loads multiple patches into the TVT which includes a transferability adaption module (TAM). Furthermore, a patch-level domain discriminator in the TAM assigns weights to the multiple patches and determines one or more transferable patches based on the weights. Moreover, the heterogeneous device generates a transferable attention output for an attention module in the TAM based on the one or more transferable patches.

Abstract: A style picture generating method, an apparatus and a non-transitory computer readable storage medium thereof are provided. The method includes: obtaining one or more models by training a neural network; obtaining a plurality of interpolated models based on the one or more models; generating a plurality of pictures by the plurality of interpolated models; and generating the style picture by combining two or more pictures in the plurality of pictures using one or more model-specific alpha masks.