Search Results for author:
Found 30 papers, 5 papers with code
FRDiff : Feature Reuse for Universal Training-free Acceleration of Diffusion Models
The substantial computational costs of diffusion models, especially due to the repeated denoising steps necessary for high-quality image generation, present a major obstacle to their widespread adoption.
Temporal Dynamic Quantization for Diffusion Models
The diffusion model has gained popularity in vision applications due to its remarkable generative performance and versatility.
Zero-Shot Learning of a Conditional Generative Adversarial Network for Data-Free Network Quantization
We propose a novel method for training a conditional generative adversarial network (CGAN) without the use of training data, called zero-shot learning of a CGAN (ZS-CGAN).
Dual-Teacher Class-Incremental Learning With Data-Free Generative Replay
In the conventional generative replay, the generative model is pre-trained for old data and shared in extra memory for later incremental learning.
Towards Fair Federated Learning with Zero-Shot Data Augmentation
Federated learning has emerged as an important distributed learning paradigm, where a server aggregates a global model from many client-trained models while having no access to the client data.
Data-Free Network Quantization With Adversarial Knowledge Distillation
The synthetic data are generated from a generator, while no data are used in training the generator and in quantization.
NTIRE 2020 Challenge on Real Image Denoising: Dataset, Methods and Results
This challenge is based on a newly collected validation and testing image datasets, and hence, named SIDD+.
NTIRE 2020 Challenge on Real-World Image Super-Resolution: Methods and Results
This paper reviews the NTIRE 2020 challenge on real world super-resolution.
GSANet: Semantic Segmentation with Global and Selective Attention
The proposed Global and Selective Attention Network (GSANet) features Atrous Spatial Pyramid Pooling (ASPP) with a novel sparsemax global attention and a novel selective attention that deploys a condensation and diffusion mechanism to aggregate the multi-scale contextual information from the extracted deep features.
HyperCon: Image-To-Video Model Transfer for Video-To-Video Translation Tasks
To combine the benefits of image and video models, we propose an image-to-video model transfer method called Hyperconsistency (HyperCon) that transforms any well-trained image model into a temporally consistent video model without fine-tuning.
End-to-End Multi-Task Denoising for the Joint Optimization of Perceptual Speech Metrics
Second, three loss functions based on SDR, PESQ and STOI are proposed to minimize the metric mismatch.
Sound Audio and Speech Processing
T-GSA: Transformer with Gaussian-weighted self-attention for speech enhancement
Transformer neural networks (TNN) demonstrated state-of-art performance on many natural language processing (NLP) tasks, replacing recurrent neural networks (RNNs), such as LSTMs or GRUs.
Audio and Speech Processing Sound
Wyner VAE: A Variational Autoencoder with Succinct Common Representation Learning
A new variational autoencoder (VAE) model is proposed that learns a succinct common representation of two correlated data variables for conditional and joint generation tasks.
Variable Rate Deep Image Compression With a Conditional Autoencoder
Our model also shows comparable and sometimes better performance than the state-of-the-art learned image compression models that deploy multiple networks trained for varying rates.
TW-SMNet: Deep Multitask Learning of Tele-Wide Stereo Matching
In this paper, we introduce the problem of estimating the real world depth of elements in a scene captured by two cameras with different field of views, where the first field of view (FOV) is a Wide FOV (WFOV) captured by a wide angle lens, and the second FOV is contained in the first FOV and is captured by a tele zoom lens.
Deep Robust Single Image Depth Estimation Neural Network Using Scene Understanding
We introduce two different scene understanding modules based on scene classification and coarse depth estimation respectively.
Learning with Succinct Common Representation Based on Wyner's Common Information
A new bimodal generative model is proposed for generating conditional and joint samples, accompanied with a training method with learning a succinct bottleneck representation.
AMNet: Deep Atrous Multiscale Stereo Disparity Estimation Networks
A stacked atrous multiscale network is proposed to aggregate rich multiscale contextual information from the cost volume which allows for estimating the disparity with high accuracy at multiple scales.
Jointly Sparse Convolutional Neural Networks in Dual Spatial-Winograd Domains
We consider the optimization of deep convolutional neural networks (CNNs) such that they provide good performance while having reduced complexity if deployed on either conventional systems with spatial-domain convolution or lower-complexity systems designed for Winograd convolution.
End-to-End Multi-Task Denoising for joint SDR and PESQ Optimization
First, the network optimization is performed on the time-domain signals after ISTFT to avoid spectrum mismatch.
DN-ResNet: Efficient Deep Residual Network for Image Denoising
The results show that DN-ResNets are more efficient, robust, and perform better denoising than current state of art deep learning methods, as well as the popular variants of the BM3D algorithm, in cases of blind and non-blind denoising of images corrupted with Poisson, Gaussian or Poisson-Gaussian noise.
Learning Sparse Low-Precision Neural Networks With Learnable Regularization
In training low-precision networks, gradient descent in the backward pass is performed with high-precision weights while quantized low-precision weights and activations are used in the forward pass to calculate the loss function for training.
Compression of Deep Convolutional Neural Networks under Joint Sparsity Constraints
In particular, the proposed framework produces one compressed model whose convolutional filters can be made sparse either in the spatial domain or in the Winograd domain.
Fused Deep Neural Networks for Efficient Pedestrian Detection
The classification system further classifies the generated candidates based on opinions of multiple deep verification networks and a fusion network which utilizes a novel soft-rejection fusion method to adjust the confidence in the detection results.
Universal Deep Neural Network Compression
In this paper, we investigate lossy compression of deep neural networks (DNNs) by weight quantization and lossless source coding for memory-efficient deployment.
CT-SRCNN: Cascade Trained and Trimmed Deep Convolutional Neural Networks for Image Super Resolution
We propose methodologies to train highly accurate and efficient deep convolutional neural networks (CNNs) for image super resolution (SR).
BridgeNets: Student-Teacher Transfer Learning Based on Recursive Neural Networks and its Application to Distant Speech Recognition
Despite the remarkable progress achieved on automatic speech recognition, recognizing far-field speeches mixed with various noise sources is still a challenging task.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+4
Residual LSTM: Design of a Deep Recurrent Architecture for Distant Speech Recognition
The residual LSTM provides an additional spatial shortcut path from lower layers for efficient training of deep networks with multiple LSTM layers.
Towards the Limit of Network Quantization
Network quantization is one of network compression techniques to reduce the redundancy of deep neural networks.
Fused DNN: A deep neural network fusion approach to fast and robust pedestrian detection
A single shot deep convolutional network is trained as a object detector to generate all possible pedestrian candidates of different sizes and occlusions.
Ranked #21 on
Pedestrian Detection
on Caltech
