Search Results for author:
Found 52 papers, 32 papers with code
PassionSR: Post-Training Quantization with Adaptive Scale in One-Step Diffusion based Image Super-Resolution
Diffusion-based image super-resolution (SR) models have shown superior performance at the cost of multiple denoising steps.
Grounding-IQA: Multimodal Language Grounding Model for Image Quality Assessment
Experiments demonstrate that our proposed task paradigm, dataset, and benchmark facilitate the more fine-grained IQA application.
MambaIRv2: Attentive State Space Restoration
The Mamba-based image restoration backbones have recently demonstrated significant potential in balancing global reception and computational efficiency.
BiDense: Binarization for Dense Prediction
BiDense incorporates two key techniques: the Distribution-adaptive Binarizer (DAB) and the Channel-adaptive Full-precision Bypass (CFB).
Hi-Mamba: Hierarchical Mamba for Efficient Image Super-Resolution
State Space Models (SSM), such as Mamba, have shown strong representation ability in modeling long-range dependency with linear complexity, achieving successful applications from high-level to low-level vision tasks.
Gap Preserving Distillation by Building Bidirectional Mappings with A Dynamic Teacher
To address this, we propose a Gap Preserving Distillation (GPD) method that trains an additional dynamic teacher model from scratch along with training the student to bridge this gap.
Distillation-Free One-Step Diffusion for Real-World Image Super-Resolution
However, these methods incur substantial training costs and may constrain the performance of the student model by the teacher's limitations.
Effective Diffusion Transformer Architecture for Image Super-Resolution
In this work, we design an effective diffusion transformer for image super-resolution (DiT-SR) that achieves the visual quality of prior-based methods, but through a training-from-scratch manner.
One Step Diffusion-based Super-Resolution with Time-Aware Distillation
Specifically, we introduce a novel score distillation strategy to align the data distribution between the outputs of the student and teacher models after minor noise perturbation.
Enhanced Long-Tailed Recognition with Contrastive CutMix Augmentation
Real-world data often follows a long-tailed distribution, where a few head classes occupy most of the data and a large number of tail classes only contain very limited samples.
Ranked #6 on
Long-tail Learning
on CIFAR-10-LT (ρ=50)
UltraPixel: Advancing Ultra-High-Resolution Image Synthesis to New Peaks
Ultra-high-resolution image generation poses great challenges, such as increased semantic planning complexity and detail synthesis difficulties, alongside substantial training resource demands.
2DQuant: Low-bit Post-Training Quantization for Image Super-Resolution
Low-bit quantization has become widespread for compressing image super-resolution (SR) models for edge deployment, which allows advanced SR models to enjoy compact low-bit parameters and efficient integer/bitwise constructions for storage compression and inference acceleration, respectively.
Binarized Diffusion Model for Image Super-Resolution
Nonetheless, due to the model structure and the multi-step iterative attribute of DMs, existing binarization methods result in significant performance degradation.
Benchmarking the Robustness of Temporal Action Detection Models Against Temporal Corruptions
In this paper, we extensively analyze the robustness of seven leading TAD methods and obtain some interesting findings: 1) Existing methods are particularly vulnerable to temporal corruptions, and end-to-end methods are often more susceptible than those with a pre-trained feature extractor; 2) Vulnerability mainly comes from localization error rather than classification error; 3) When corruptions occur in the middle of an action instance, TAD models tend to yield the largest performance drop.
CustomVideo: Customizing Text-to-Video Generation with Multiple Subjects
Customized text-to-video generation aims to generate high-quality videos guided by text prompts and subject references.
CoDe: An Explicit Content Decoupling Framework for Image Restoration
To address this issue we propose an explicit Content Decoupling framework for IR dubbed CoDe to end-to-end model the restoration process by utilizing decoupled content components in a divide-and-conquer-like architecture.
Bridging Code Semantic and LLMs: Semantic Chain-of-Thought Prompting for Code Generation
Meanwhile, while traditional techniques leveraging such semantic information require complex static or dynamic code analysis to obtain features such as data flow and control flow, SeCoT demonstrates that this process can be fully automated via the intrinsic capabilities of LLMs (i. e., in-context learning), while being generalizable and applicable to challenging domains.
DriveGPT4: Interpretable End-to-end Autonomous Driving via Large Language Model
Multimodal large language models (MLLMs) have emerged as a prominent area of interest within the research community, given their proficiency in handling and reasoning with non-textual data, including images and videos.
Robustifying Token Attention for Vision Transformers
Interestingly, we observe that the attention mechanism of ViTs tends to rely on few important tokens, a phenomenon we call token overfocusing.
Improving Robustness of Vision Transformers by Reducing Sensitivity To Patch Corruptions
Despite their success, vision transformers still remain vulnerable to image corruptions, such as noise or blur.
Boosting Semi-Supervised Learning with Contrastive Complementary Labeling
In this way, these data can be also very informative if we can effectively exploit these complementary labels, i. e., the classes that a sample does not belong to.
Downscaled Representation Matters: Improving Image Rescaling with Collaborative Downscaled Images
Inspired by this, we propose a Hierarchical Collaborative Downscaling (HCD) method that performs gradient descent in both HR and LR domains to improve the downscaled representations.
Automated Dominative Subspace Mining for Efficient Neural Architecture Search
Then, we perform a local search within the mined subspace to find effective architectures.
Ranked #2 on
Neural Architecture Search
on NAS-Bench-201, ImageNet-16-120
(Accuracy (Val) metric)
Pareto-aware Neural Architecture Generation for Diverse Computational Budgets
More critically, these independent search processes cannot share their learned knowledge (i. e., the distribution of good architectures) with each other and thus often result in limited search results.
Improving Fine-tuning of Self-supervised Models with Contrastive Initialization
Self-supervised learning (SSL) has achieved remarkable performance in pretraining the models that can be further used in downstream tasks via fine-tuning.
Towards Lightweight Super-Resolution with Dual Regression Learning
Nevertheless, it is hard for existing model compression methods to accurately identify the redundant components due to the extremely large SR mapping space.
Improving Robustness by Enhancing Weak Subnets
We show that EWS greatly improves both robustness against corrupted images as well as accuracy on clean data.
Improvement of image classification by multiple optical scattering
Multiple optical scattering occurs when light propagates in a non-uniform medium.
AdaXpert: Adapting Neural Architecture for Growing Data
To address this, we present a neural architecture adaptation method, namely Adaptation eXpert (AdaXpert), to efficiently adjust previous architectures on the growing data.
Content-Aware Convolutional Neural Networks
In practice, the convolutional operation on some of the windows (e. g., smooth windows that contain very similar pixels) can be very redundant and may introduce noises into the computation.
Contrastive Neural Architecture Search with Neural Architecture Comparators
One of the key steps in Neural Architecture Search (NAS) is to estimate the performance of candidate architectures.
Pareto-Frontier-aware Neural Architecture Generation for Diverse Budgets
To this end, we propose a Pareto-Frontier-aware Neural Architecture Generator (NAG) which takes an arbitrary budget as input and produces the Pareto optimal architecture for the target budget.
Towards Accurate and Compact Architectures via Neural Architecture Transformer
To address this issue, we propose a Neural Architecture Transformer++ (NAT++) method which further enlarges the set of candidate transitions to improve the performance of architecture optimization.
Deep View Synthesis via Self-Consistent Generative Network
To address the above issues, in this paper, we propose a novel deep generative model, called Self-Consistent Generative Network (SCGN), which synthesizes novel views from the given input views without explicitly exploiting the geometric information.
Pareto-Frontier-aware Neural Architecture Search
To find promising architectures under different budgets, existing methods may have to perform an independent search for each budget, which is very inefficient and unnecessary.
Double Forward Propagation for Memorized Batch Normalization
Although the standard BN can significantly accelerate the training of DNNs and improve the generalization performance, it has several underlying limitations which may hamper the performance in both training and inference.
Conditional Automated Channel Pruning for Deep Neural Networks
To address this issue, given a target compression rate for the whole model, one can search for the optimal compression rate for each layer.
Improving Generative Adversarial Networks with Local Coordinate Coding
In this paper, rather than sampling from the predefined prior distribution, we propose an LCCGAN model with local coordinate coding (LCC) to improve the performance of generating data.
Breaking the Curse of Space Explosion: Towards Efficient NAS with Curriculum Search
With the proposed search strategy, our Curriculum Neural Architecture Search (CNAS) method significantly improves the search efficiency and finds better architectures than existing NAS methods.
Disturbance-immune Weight Sharing for Neural Architecture Search
To alleviate the performance disturbance issue, we propose a new disturbance-immune update strategy for model updating.
Closed-loop Matters: Dual Regression Networks for Single Image Super-Resolution
Extensive experiments with paired training data and unpaired real-world data demonstrate our superiority over existing methods.
Hierarchical Neural Architecture Search for Single Image Super-Resolution
To this end, we design a hierarchical SR search space and propose a hierarchical controller for architecture search.
Joint Wasserstein Distribution Matching
However, the resultant optimization problem is still intractable.
Discrimination-aware Network Pruning for Deep Model Compression
In this paper, we propose a simple-yet-effective method called discrimination-aware channel pruning (DCP) to choose the channels that actually contribute to the discriminative power.
NAT: Neural Architecture Transformer for Accurate and Compact Architectures
To verify the effectiveness of the proposed strategies, we apply NAT on both hand-crafted architectures and NAS based architectures.
Auto-Embedding Generative Adversarial Networks for High Resolution Image Synthesis
To address this issue, we develop a novel GAN called Auto-Embedding Generative Adversarial Network (AEGAN), which simultaneously encodes the global structure features and captures the fine-grained details.
Discrimination-aware Channel Pruning for Deep Neural Networks
Channel pruning is one of the predominant approaches for deep model compression.
Learning Joint Wasserstein Auto-Encoders for Joint Distribution Matching
We study the joint distribution matching problem which aims at learning bidirectional mappings to match the joint distribution of two domains.
Open-Ended Question Answering
Unsupervised Image-To-Image Translation
+2
Dual Reconstruction Nets for Image Super-Resolution with Gradient Sensitive Loss
However, most deep learning methods employ feed-forward architectures, and thus the dependencies between LR and HR images are not fully exploited, leading to limited learning performance.
Adversarial Learning with Local Coordinate Coding
Generative adversarial networks (GANs) aim to generate realistic data from some prior distribution (e. g., Gaussian noises).
The Shallow End: Empowering Shallower Deep-Convolutional Networks through Auxiliary Outputs
As a result, the representation power of intermediate layers can be very weak and the model becomes very redundant with limited performance.
High-speed real-time single-pixel microscopy based on Fourier sampling
However, according to CS theory, image reconstruction is an iterative process that consumes enormous amounts of computational time and cannot be performed in real time.
