Search Results for author:
Found 130 papers, 80 papers with code
Parts for the Whole: The DCT Norm for Extreme Visual Recovery
Here we study the extreme visual recovery problem, in which over 90\% of pixel values in a given image are missing.
Streaming Label Learning for Modeling Labels on the Fly
The core of SLL is to explore and exploit the relationships between new labels and past labels and then inherit the relationship into hypotheses of labels to boost the performance of new classifiers.
CNNpack: Packing Convolutional Neural Networks in the Frequency Domain
Deep convolutional neural networks (CNNs) are successfully used in a number of applications.
Privileged Multi-label Learning
This paper presents privileged multi-label learning (PrML) to explore and exploit the relationship between labels in multi-label learning problems.
Towards Evolutional Compression
In contrast to directly recognizing subtle weights or filters as redundant in a given CNN, this paper presents an evolutionary method to automatically eliminate redundant convolution filters.
Beyond Filters: Compact Feature Map for Portable Deep Model
The filter is then re-configured to establish the mapping from original input to the new compact feature map, and the resulting network can preserve intrinsic information of the original network with significantly fewer parameters, which not only decreases the online memory for launching CNN but also accelerates the computation speed.
AutoEncoder Inspired Unsupervised Feature Selection
High-dimensional data in many areas such as computer vision and machine learning tasks brings in computational and analytical difficulty.
Learning Versatile Filters for Efficient Convolutional Neural Networks
A series of secondary filters can be derived from a primary filter.
Learning Student Networks via Feature Embedding
Experiments on benchmark datasets and well-trained networks suggest that the proposed algorithm is superior to state-of-the-art teacher-student learning methods in terms of computational and storage complexity.
Data-Free Learning of Student Networks
Learning portable neural networks is very essential for computer vision for the purpose that pre-trained heavy deep models can be well applied on edge devices such as mobile phones and micro sensors.
Co-Evolutionary Compression for Unpaired Image Translation
Generative adversarial networks (GANs) have been successfully used for considerable computer vision tasks, especially the image-to-image translation.
Learning Instance-wise Sparsity for Accelerating Deep Models
Exploring deep convolutional neural networks of high efficiency and low memory usage is very essential for a wide variety of machine learning tasks.
Attribute Aware Pooling for Pedestrian Attribute Recognition
This paper expands the strength of deep convolutional neural networks (CNNs) to the pedestrian attribute recognition problem by devising a novel attribute aware pooling algorithm.
Full-Stack Filters to Build Minimum Viable CNNs
Existing works used to decrease the number or size of requested convolution filters for a minimum viable CNN on edge devices.
CARS: Continuous Evolution for Efficient Neural Architecture Search
Architectures in the population that share parameters within one SuperNet in the latest generation will be tuned over the training dataset with a few epochs.
Searching for Accurate Binary Neural Architectures
Binary neural networks have attracted tremendous attention due to the efficiency for deploying them on mobile devices.
Positive-Unlabeled Compression on the Cloud
In practice, only a small portion of the original training set is required as positive examples and more useful training examples can be obtained from the massive unlabeled data on the cloud through a PU classifier with an attention based multi-scale feature extractor.
Efficient Residual Dense Block Search for Image Super-Resolution
Focusing on this issue, we propose an efficient residual dense block search algorithm with multiple objectives to hunt for fast, lightweight and accurate networks for image super-resolution.
ReNAS:Relativistic Evaluation of Neural Architecture Search
An effective and efficient architecture performance evaluation scheme is essential for the success of Neural Architecture Search (NAS).
GhostNet: More Features from Cheap Operations
Deploying convolutional neural networks (CNNs) on embedded devices is difficult due to the limited memory and computation resources.
Ranked #864 on
Image Classification
on ImageNet
AdderNet: Do We Really Need Multiplications in Deep Learning?
The widely-used convolutions in deep neural networks are exactly cross-correlation to measure the similarity between input feature and convolution filters, which involves massive multiplications between float values.
Widening and Squeezing: Towards Accurate and Efficient QNNs
Quantization neural networks (QNNs) are very attractive to the industry because their extremely cheap calculation and storage overhead, but their performance is still worse than that of networks with full-precision parameters.
On Positive-Unlabeled Classification in GAN
In contrast, it is more reasonable to treat the generated data as unlabeled, which could be positive or negative according to their quality.
Discernible Image Compression
Experiments on benchmarks demonstrate that images compressed by using the proposed method can also be well recognized by subsequent visual recognition and detection models.
Beyond Dropout: Feature Map Distortion to Regularize Deep Neural Networks
On one hand, massive trainable parameters significantly enhance the performance of these deep networks.
Automatically Searching for U-Net Image Translator Architecture
Moreover, we transplant the searched network architecture to other datasets which are not involved in the architecture searching procedure.
Distilling portable Generative Adversarial Networks for Image Translation
To promote the capability of student generator, we include a student discriminator to measure the distances between real images, and images generated by student and teacher generators.
Hit-Detector: Hierarchical Trinity Architecture Search for Object Detection
To this end, we propose a hierarchical trinity search framework to simultaneously discover efficient architectures for all components (i. e. backbone, neck, and head) of object detector in an end-to-end manner.
A Semi-Supervised Assessor of Neural Architectures
A graph convolutional neural network is introduced to predict the performance of architectures based on the learned representations and their relation modeled by the graph.
DC-NAS: Divide-and-Conquer Neural Architecture Search
Neural architecture searching is a way of automatically exploring optimal deep neural networks in a given huge search space.
HourNAS: Extremely Fast Neural Architecture Search Through an Hourglass Lens
To achieve an extremely fast NAS while preserving the high accuracy, we propose to identify the vital blocks and make them the priority in the architecture search.
Optical Flow Distillation: Towards Efficient and Stable Video Style Transfer
This paper proposes to learn a lightweight video style transfer network via knowledge distillation paradigm.
MTP: Multi-Task Pruning for Efficient Semantic Segmentation Networks
To identify the redundancy in segmentation networks, we present a multi-task channel pruning approach.
Adversarially Robust Neural Architectures
We explore the relationship among adversarial robustness, Lipschitz constant, and architecture parameters and show that an appropriate constraint on architecture parameters could reduce the Lipschitz constant to further improve the robustness.
Searching for Low-Bit Weights in Quantized Neural Networks
Quantized neural networks with low-bit weights and activations are attractive for developing AI accelerators.
AdderSR: Towards Energy Efficient Image Super-Resolution
To this end, we thoroughly analyze the relationship between an adder operation and the identity mapping and insert shortcuts to enhance the performance of SR models using adder networks.
Kernel Based Progressive Distillation for Adder Neural Networks
A convolutional neural network (CNN) with the same architecture is simultaneously initialized and trained as a teacher network, features and weights of ANN and CNN will be transformed to a new space to eliminate the accuracy drop.
Training Binary Neural Networks through Learning with Noisy Supervision
This paper formalizes the binarization operations over neural networks from a learning perspective.
SCOP: Scientific Control for Reliable Neural Network Pruning
To increase the reliability of the results, we prefer to have a more rigorous research design by including a scientific control group as an essential part to minimize the effect of all factors except the association between the filter and expected network output.
Model Rubik's Cube: Twisting Resolution, Depth and Width for TinyNets
To this end, we summarize a tiny formula for downsizing neural architectures through a series of smaller models derived from the EfficientNet-B0 with the FLOPs constraint.
Ranked #692 on
Image Classification
on ImageNet
VEGA: Towards an End-to-End Configurable AutoML Pipeline
Automated Machine Learning (AutoML) is an important industrial solution for automatic discovery and deployment of the machine learning models.
Dynamic Feature Pyramid Networks for Object Detection
An attempt to enhance the FPN is enriching the spatial information by expanding the receptive fields, which is promising to largely improve the detection accuracy.
Model Rubik’s Cube: Twisting Resolution, Depth and Width for TinyNets
To this end, we summarize a tiny formula for downsizing neural architectures through a series of smaller models derived from the EfficientNet-B0 with the FLOPs constraint.
Adapting Neural Architectures Between Domains
The power of deep neural networks is to be unleashed for analyzing a large volume of data (e. g. ImageNet), but the architecture search is often executed on another smaller dataset (e. g. CIFAR-10) to finish it in a feasible time.
Pre-Trained Image Processing Transformer
To maximally excavate the capability of transformer, we present to utilize the well-known ImageNet benchmark for generating a large amount of corrupted image pairs.
Ranked #1 on
Single Image Deraining
on Rain100L
(using extra training data)
Residual Distillation: Towards Portable Deep Neural Networks without Shortcuts
In particular, we propose a novel joint-training framework to train plain CNN by leveraging the gradients of the ResNet counterpart.
A Survey on Visual Transformer
Transformer, first applied to the field of natural language processing, is a type of deep neural network mainly based on the self-attention mechanism.
GhostSR: Learning Ghost Features for Efficient Image Super-Resolution
Based on the observation that many features in SISR models are also similar to each other, we propose to use shift operation to generate the redundant features (i. e., ghost features).
AdderNet and its Minimalist Hardware Design for Energy-Efficient Artificial Intelligence
With a comprehensive comparison on the performance, power consumption, hardware resource consumption and network generalization capability, we conclude the AdderNet is able to surpass all the other competitors including the classical CNN, novel memristor-network, XNOR-Net and the shift-kernel based network, indicating its great potential in future high performance and energy-efficient artificial intelligence applications.
Transformer in Transformer
In this paper, we point out that the attention inside these local patches are also essential for building visual transformers with high performance and we explore a new architecture, namely, Transformer iN Transformer (TNT).
Learning Frequency Domain Approximation for Binary Neural Networks
Binary neural networks (BNNs) represent original full-precision weights and activations into 1-bit with sign function.
Manifold Regularized Dynamic Network Pruning
Then, the manifold relationship between instances and the pruned sub-networks will be aligned in the training procedure.
Learning Frequency-aware Dynamic Network for Efficient Super-Resolution
Extensive experiments conducted on benchmark SISR models and datasets show that the frequency-aware dynamic network can be employed for various SISR neural architectures to obtain the better tradeoff between visual quality and computational complexity.
Distilling Object Detectors via Decoupled Features
To this end, we present a novel distillation algorithm via decoupled features (DeFeat) for learning a better student detector.
Coarse-to-Fine Searching for Efficient Generative Adversarial Networks
This paper studies the neural architecture search (NAS) problem for developing efficient generator networks.
Winograd Algorithm for AdderNet
Adder neural network (AdderNet) is a new kind of deep model that replaces the original massive multiplications in convolutions by additions while preserving the high performance.
Universal Adder Neural Networks
The widely-used convolutions in deep neural networks are exactly cross-correlation to measure the similarity between input feature and convolution filters, which involves massive multiplications between float values.
Patch Slimming for Efficient Vision Transformers
We first identify the effective patches in the last layer and then use them to guide the patch selection process of previous layers.
Ranked #8 on
Efficient ViTs
on ImageNet-1K (with DeiT-T)
Dynamic Resolution Network
To this end, we propose a novel dynamic-resolution network (DRNet) in which the input resolution is determined dynamically based on each input sample.
Data-Free Knowledge Distillation for Image Super-Resolution
Experiments on various datasets and architectures demonstrate that the proposed method is able to be utilized for effectively learning portable student networks without the original data, e. g., with 0. 16dB PSNR drop on Set5 for x2 super resolution.
Positive-Unlabeled Data Purification in the Wild for Object Detection
In this paper, we present a positive-unlabeled learning based scheme to expand training data by purifying valuable images from massive unlabeled ones, where the original training data are viewed as positive data and the unlabeled images in the wild are unlabeled data.
Learning Student Networks in the Wild
Experiments on various datasets demonstrate that the student networks learned by the proposed method can achieve comparable performance with those using the original dataset.
ReNAS: Relativistic Evaluation of Neural Architecture Search
An effective and efficient architecture performance evaluation scheme is essential for the success of Neural Architecture Search (NAS).
Federated Learning with Positive and Unlabeled Data
We study the problem of learning from positive and unlabeled (PU) data in the federated setting, where each client only labels a little part of their dataset due to the limitation of resources and time.
Post-Training Quantization for Vision Transformer
Recently, transformer has achieved remarkable performance on a variety of computer vision applications.
Augmented Shortcuts for Vision Transformers
Transformer models have achieved great progress on computer vision tasks recently.
Learning Efficient Vision Transformers via Fine-Grained Manifold Distillation
Specifically, we train a tiny student model to match a pre-trained teacher model in the patch-level manifold space.
CMT: Convolutional Neural Networks Meet Vision Transformers
Vision transformers have been successfully applied to image recognition tasks due to their ability to capture long-range dependencies within an image.
Greedy Network Enlarging
Recent studies on deep convolutional neural networks present a simple paradigm of architecture design, i. e., models with more MACs typically achieve better accuracy, such as EfficientNet and RegNet.
Neural Architecture Dilation for Adversarial Robustness
With the tremendous advances in the architecture and scale of convolutional neural networks (CNNs) over the past few decades, they can easily reach or even exceed the performance of humans in certain tasks.
Hire-MLP: Vision MLP via Hierarchical Rearrangement
Previous vision MLPs such as MLP-Mixer and ResMLP accept linearly flattened image patches as input, making them inflexible for different input sizes and hard to capture spatial information.
Learning Versatile Convolution Filters for Efficient Visual Recognition
A series of secondary filters can be derived from a primary filter with the help of binary masks.
Positive and Unlabeled Federated Learning
We study the problem of learning from positive and unlabeled (PU) data in the federated setting, where each client only labels a little part of their dataset due to the limitation of resources and time.
A Self-adaptive Weighted Differential Evolution Approach for Large-scale Feature Selection
We demonstrate the effectiveness of our algorithm on twelve large-scale datasets.
An Image Patch is a Wave: Phase-Aware Vision MLP
To dynamically aggregate tokens, we propose to represent each token as a wave function with two parts, amplitude and phase.
Towards Stable and Robust AdderNets
Adder neural network (AdderNet) replaces the original convolutions with massive multiplications by cheap additions while achieving comparable performance thus yields a series of energy-efficient neural networks.
Handling Long-tailed Feature Distribution in AdderNets
Adder neural networks (ANNs) are designed for low energy cost which replace expensive multiplications in convolutional neural networks (CNNs) with cheaper additions to yield energy-efficient neural networks and hardware accelerations.
Adder Attention for Vision Transformer
With the new operation, vision transformers constructed using additions can also provide powerful feature representations.
An Empirical Study of Adder Neural Networks for Object Detection
Adder neural networks (AdderNets) have shown impressive performance on image classification with only addition operations, which are more energy efficient than traditional convolutional neural networks built with multiplications.
Instance-Aware Dynamic Neural Network Quantization
However, natural images are of huge diversity with abundant content and using such a universal quantization configuration for all samples is not an optimal strategy.
PyramidTNT: Improved Transformer-in-Transformer Baselines with Pyramid Architecture
We hope this new baseline will be helpful to the further research and application of vision transformer.
GhostNets on Heterogeneous Devices via Cheap Operations
The proposed C-Ghost module can be taken as a plug-and-play component to upgrade existing convolutional neural networks.
DropNAS: Grouped Operation Dropout for Differentiable Architecture Search
Neural architecture search (NAS) has shown encouraging results in automating the architecture design.
Brain-inspired Multilayer Perceptron with Spiking Neurons
However, due to the simplicity of their structures, the performance highly depends on the local features communication machenism.
MaskGroup: Hierarchical Point Grouping and Masking for 3D Instance Segmentation
For instance, our approach achieves a 66. 4\% mAP with the 0. 5 IoU threshold on the ScanNetV2 test set, which is 1. 9\% higher than the state-of-the-art method.
Ranked #6 on
3D Instance Segmentation
on S3DIS
Multimodal Token Fusion for Vision Transformers
Many adaptations of transformers have emerged to address the single-modal vision tasks, where self-attention modules are stacked to handle input sources like images.
Ranked #1 on
Semantic Segmentation
on SUN-RGBD
(using extra training data)
Source-Free Domain Adaptation via Distribution Estimation
Domain Adaptation aims to transfer the knowledge learned from a labeled source domain to an unlabeled target domain whose data distributions are different.
Vision GNN: An Image is Worth Graph of Nodes
In this paper, we propose to represent the image as a graph structure and introduce a new Vision GNN (ViG) architecture to extract graph-level feature for visual tasks.
Ranked #362 on
Image Classification
on ImageNet
Network Amplification With Efficient MACs Allocation
In particular, the proposed method is used to enlarge models sourced by GhostNet, we achieve state-of-the-art 80. 9% and 84. 3% ImageNet top-1 accuracies under the setting of 600M and 4. 4B MACs, respectively.
Spatial-Channel Token Distillation for Vision MLPs
Recently, neural architectures with all Multi-layer Perceptrons (MLPs) have attracted great research interest from the computer vision community.
Ranked #488 on
Image Classification
on ImageNet
AdaBin: Improving Binary Neural Networks with Adaptive Binary Sets
Since the modern deep neural networks are of sophisticated design with complex architecture for the accuracy reason, the diversity on distributions of weights and activations is very high.
Bridge the Gap Between Architecture Spaces via A Cross-Domain Predictor
To solve this issue, we propose a Cross-Domain Predictor (CDP), which is trained based on the existing NAS benchmark datasets (e. g., NAS-Bench-101), but can be used to find high-performance architectures in large-scale search spaces.
GhostNetV2: Enhance Cheap Operation with Long-Range Attention
The convolutional operation can only capture local information in a window region, which prevents performance from being further improved.
FastMIM: Expediting Masked Image Modeling Pre-training for Vision
This paper presents FastMIM, a simple and generic framework for expediting masked image modeling with the following two steps: (i) pre-training vision backbones with low-resolution input images; and (ii) reconstructing Histograms of Oriented Gradients (HOG) feature instead of original RGB values of the input images.
Redistribution of Weights and Activations for AdderNet Quantization
To this end, we first thoroughly analyze the difference on distributions of weights and activations in AdderNet and then propose a new quantization algorithm by redistributing the weights and the activations.
BiMLP: Compact Binary Architectures for Vision Multi-Layer Perceptrons
This paper studies the problem of designing compact binary architectures for vision multi-layer perceptrons (MLPs).
Network Expansion for Practical Training Acceleration
Recently, the sizes of deep neural networks and training datasets both increase drastically to pursue better performance in a practical sense.
RefSR-NeRF: Towards High Fidelity and Super Resolution View Synthesis
We present Reference-guided Super-Resolution Neural Radiance Field (RefSR-NeRF) that extends NeRF to super resolution and photorealistic novel view synthesis.
Toward Accurate Post-Training Quantization for Image Super Resolution
In this paper, we study post-training quantization(PTQ) for image super resolution using only a few unlabeled calibration images.
Masked Image Modeling with Local Multi-Scale Reconstruction
The lower layers are not explicitly guided and the interaction among their patches is only used for calculating new activations.
VanillaNet: the Power of Minimalism in Deep Learning
In this study, we introduce VanillaNet, a neural network architecture that embraces elegance in design.
VanillaKD: Revisit the Power of Vanilla Knowledge Distillation from Small Scale to Large Scale
The tremendous success of large models trained on extensive datasets demonstrates that scale is a key ingredient in achieving superior results.
Multiscale Positive-Unlabeled Detection of AI-Generated Texts
Recent releases of Large Language Models (LLMs), e. g. ChatGPT, are astonishing at generating human-like texts, but they may impact the authenticity of texts.
GPT4Image: Can Large Pre-trained Models Help Vision Models on Perception Tasks?
We present a new learning paradigm in which the knowledge extracted from large pre-trained models are utilized to help models like CNN and ViT learn enhanced representations and achieve better performance.
ParameterNet: Parameters Are All You Need
In the language domain, LLaMA-1B enhanced with ParameterNet achieves 2\% higher accuracy over vanilla LLaMA.
Less is More: Focus Attention for Efficient DETR
DETR-like models have significantly boosted the performance of detectors and even outperformed classical convolutional models.
Category Feature Transformer for Semantic Segmentation
Aggregation of multi-stage features has been revealed to play a significant role in semantic segmentation.
Gold-YOLO: Efficient Object Detector via Gather-and-Distribute Mechanism
In the past years, YOLO-series models have emerged as the leading approaches in the area of real-time object detection.
IFT: Image Fusion Transformer for Ghost-free High Dynamic Range Imaging
By matching similar patches between frames, objects with large motion ranges in dynamic scenes can be aligned, which can effectively alleviate the generation of artifacts.
One-for-All: Bridge the Gap Between Heterogeneous Architectures in Knowledge Distillation
To tackle the challenge in distilling heterogeneous models, we propose a simple yet effective one-for-all KD framework called OFA-KD, which significantly improves the distillation performance between heterogeneous architectures.
Data-Free Distillation of Language Model by Text-to-Text Transfer
Data-Free Knowledge Distillation (DFKD) plays a vital role in compressing the model when original training data is unavailable.
Towards Higher Ranks via Adversarial Weight Pruning
To this end, we propose a Rank-based PruninG (RPG) method to maintain the ranks of sparse weights in an adversarial manner.
LightCLIP: Learning Multi-Level Interaction for Lightweight Vision-Language Models
Moreover, based on the observation that the accuracy of CLIP model does not increase correspondingly as the parameters of text encoder increase, an extra objective of masked language modeling (MLM) is leveraged for maximizing the potential of the shortened text encoder.
UFineBench: Towards Text-based Person Retrieval with Ultra-fine Granularity
Firstly, we construct a new \textbf{dataset} named UFine6926.
GenDet: Towards Good Generalizations for AI-Generated Image Detection
The misuse of AI imagery can have harmful societal effects, prompting the creation of detectors to combat issues like the spread of fake news.
CBQ: Cross-Block Quantization for Large Language Models
Post-training quantization (PTQ) has played a key role in compressing large language models (LLMs) with ultra-low costs.
TinySAM: Pushing the Envelope for Efficient Segment Anything Model
Extensive experiments on various zero-shot transfer tasks demonstrate the significantly advantageous performance of our TinySAM against counterpart methods.
DECO: Query-Based End-to-End Object Detection with ConvNets
The proposed framework, \ie, Detection ConvNet (DECO), is composed of a backbone and convolutional encoder-decoder architecture.
PanGu-$π$: Enhancing Language Model Architectures via Nonlinearity Compensation
We then demonstrate that the proposed approach is significantly effective for enhancing the model nonlinearity through carefully designed ablations; thus, we present a new efficient model architecture for establishing modern, namely, PanGu-$\pi$.
An Empirical Study of Scaling Law for OCR
The laws of model size, data volume, computation and model performance have been extensively studied in the field of Natural Language Processing (NLP).
Ranked #1 on
Scene Text Recognition
on ICDAR2013
(using extra training data)
Optical Character Recognition
Optical Character Recognition (OCR)
+1
A Survey on Transformer Compression
Model compression methods reduce the memory and computational cost of Transformer, which is a necessary step to implement large language/vision models on practical devices.
Rethinking Optimization and Architecture for Tiny Language Models
Several design formulas are empirically proved especially effective for tiny language models, including tokenizer compression, architecture tweaking, parameter inheritance and multiple-round training.
Vision Superalignment: Weak-to-Strong Generalization for Vision Foundation Models
Recent advancements in large language models have sparked interest in their extraordinary and near-superhuman capabilities, leading researchers to explore methods for evaluating and optimizing these abilities, which is called superalignment.
DenseMamba: State Space Models with Dense Hidden Connection for Efficient Large Language Models
Large language models (LLMs) face a daunting challenge due to the excessive computational and memory requirements of the commonly used Transformer architecture.
SAM-DiffSR: Structure-Modulated Diffusion Model for Image Super-Resolution
In this paper, we propose the SAM-DiffSR model, which can utilize the fine-grained structure information from SAM in the process of sampling noise to improve the image quality without additional computational cost during inference.
Distilling Semantic Priors from SAM to Efficient Image Restoration Models
SPD leverages a self-distillation manner to distill the fused semantic priors to boost the performance of original IR models.
DiJiang: Efficient Large Language Models through Compact Kernelization
In an effort to reduce the computational load of Transformers, research on linear attention has gained significant momentum.
IPT-V2: Efficient Image Processing Transformer using Hierarchical Attentions
Recent advances have demonstrated the powerful capability of transformer architecture in image restoration.
Exhaustive Exploitation of Nature-inspired Computation for Cancer Screening in an Ensemble Manner
This study presents a framework termed Evolutionary Optimized Diverse Ensemble Learning (EODE) to improve ensemble learning for cancer classification from gene expression data.
Neural Architecture Search in a Proxy Validation Loss Landscape
This paper searches for the optimal neural architecture by minimizing a proxy of validation loss.
