Search Results for author:
Found 19 papers, 9 papers with code
Towards Accurate Post-training Network Quantization via Bit-Split and Stitching
Network quantization is essential for deploying deep models to IoT devices due to the high efficiency, no matter on special hardware like TPU or general hardware like CPU and GPU.
ProxyBNN: Learning Binarized Neural Networks via Proxy Matrices
The matrix composed of basis vectors is referred to as the proxy matrix, and auxiliary variables serve as the coefficients of this linear combination.
$Z^*$: Zero-shot Style Transfer via Attention Rearrangement
Despite the remarkable progress in image style transfer, formulating style in the context of art is inherently subjective and challenging.
Singular Value Fine-tuning: Few-shot Segmentation requires Few-parameters Fine-tuning
In this paper, we rethink the paradigm and explore a new regime: {\em fine-tuning a small part of parameters in the backbone}.
Ranked #8 on
Few-Shot Semantic Segmentation
on COCO-20i (1-shot)
Soft Threshold Ternary Networks
The latest STTN shows that ResNet-18 with ternary weights and ternary activations achieves up to 68. 2% Top-1 accuracy on ImageNet.
Revisiting L1 Loss in Super-Resolution: A Probabilistic View and Beyond
Super-resolution as an ill-posed problem has many high-resolution candidates for a low-resolution input.
APRIL: Finding the Achilles' Heel on Privacy for Vision Transformers
Showing how vision Transformers are at the risk of privacy leakage via gradients, we urge the significance of designing privacy-safer Transformer models and defending schemes.
Improving Binary Neural Networks through Fully Utilizing Latent Weights
Binary Neural Networks (BNNs) rely on a real-valued auxiliary variable W to help binary training.
Architecture Aware Latency Constrained Sparse Neural Networks
Acceleration of deep neural networks to meet a specific latency constraint is essential for their deployment on mobile devices.
Generative Zero-shot Network Quantization
Convolutional neural networks are able to learn realistic image priors from numerous training samples in low-level image generation and restoration.
Dynamic Dual Gating Neural Networks
In particular, dynamic dual gating can provide 59. 7% saving in computing of ResNet50 with 76. 41% top-1 accuracy on ImageNet, which has advanced the state-of-the-art.
AIM 2020 Challenge on Efficient Super-Resolution: Methods and Results
This paper reviews the AIM 2020 challenge on efficient single image super-resolution with focus on the proposed solutions and results.
Location-aware Upsampling for Semantic Segmentation
Many successful learning targets such as minimizing dice loss and cross-entropy loss have enabled unprecedented breakthroughs in segmentation tasks.
Ranked #35 on
Semantic Segmentation
on PASCAL Context
SpatialFlow: Bridging All Tasks for Panoptic Segmentation
Object location is fundamental to panoptic segmentation as it is related to all things and stuff in the image scene.
Ranked #17 on
Panoptic Segmentation
on COCO test-dev
A System-Level Solution for Low-Power Object Detection
As a case study, we evaluate our object detection system on a real-world surveillance video with input size of 512x512, and it turns out that the system can achieve an inference speed of 18 fps at the cost of 6. 9W (with display) with an mAP of 66. 4 verified on the PASCAL VOC 2012 dataset.
Compact Global Descriptor for Neural Networks
Long-range dependencies modeling, widely used in capturing spatiotemporal correlation, has shown to be effective in CNN dominated computer vision tasks.
Ranked #197 on
Object Detection
on COCO test-dev
K-Nearest Neighbors Hashing
Hashing based approximate nearest neighbor search embeds high dimensional data to compact binary codes, which enables efficient similarity search and storage.
ODE-Inspired Network Design for Single Image Super-Resolution
By casting the numerical schemes in ODE as blueprints, we derive two types of network structures: LF-block and RK-block, which correspond to the Leapfrog method and Runge-Kutta method in numerical ordinary differential equations.
Learning Compression from Limited Unlabeled Data
Through quantization or pruning, most methods may compress a large number of parameters but ignore the core role in performance degradation, which is the Gaussian conjugate prior induced by batch normalization.
