Search Results for author:
Found 51 papers, 27 papers with code
PYRA: Parallel Yielding Re-Activation for Training-Inference Efficient Task Adaptation
Consequently, a simple combination of them cannot guarantee accomplishing both training efficiency and inference efficiency with minimal costs.
Debiased Novel Category Discovering and Localization
In recent years, object detection in deep learning has experienced rapid development.
One-Dimensional Adapter to Rule Them All: Concepts, Diffusion Models and Erasing Applications
The prevalent use of commercial and open-source diffusion models (DMs) for text-to-image generation prompts risk mitigation to prevent undesired behaviors.
Re-parameterized Low-rank Prompt: Generalize a Vision-Language Model within 0.5K Parameters
On the other hand, complicated structures and update rules largely increase the computation and storage cost.
RepViT-SAM: Towards Real-Time Segmenting Anything
Here, to achieve real-time segmenting anything on mobile devices, following MobileSAM, we replace the heavyweight image encoder in SAM with RepViT model, ending up with the RepViT-SAM model.
MiLe Loss: a New Loss for Mitigating the Bias of Learning Difficulties in Generative Language Models
Experiments reveal that models incorporating the proposed MiLe Loss can gain consistent performance improvement on downstream benchmarks.
Ranked #89 on
Multi-task Language Understanding
on MMLU
Confidence-based Visual Dispersal for Few-shot Unsupervised Domain Adaptation
To address this issue, recent works consider the Few-shot Unsupervised Domain Adaptation (FUDA) where only a few source samples are labeled, and conduct knowledge transfer via self-supervised learning methods.
CAIT: Triple-Win Compression towards High Accuracy, Fast Inference, and Favorable Transferability For ViTs
We further employ a consistent dynamic channel pruning (CDCP) strategy to dynamically prune unimportant channels in ViTs.
RepViT: Revisiting Mobile CNN From ViT Perspective
Recently, lightweight Vision Transformers (ViTs) demonstrate superior performance and lower latency, compared with lightweight Convolutional Neural Networks (CNNs), on resource-constrained mobile devices.
Consolidator: Mergeable Adapter with Grouped Connections for Visual Adaptation
To further enhance the model's capacity to transfer knowledge under a constrained storage budget and keep inference efficient, we consolidate the parameters in two stages: 1. between adaptation and storage, and 2. between loading and inference.
Box-Level Active Detection
Active learning selects informative samples for annotation within budget, which has proven efficient recently on object detection.
X-ReID: Cross-Instance Transformer for Identity-Level Person Re-Identification
In this paper, we propose to promote Instance-Level features to Identity-Level features by employing cross-attention to incorporate information from one image to another of the same identity, thus more unified and discriminative pedestrian information can be obtained.
Confidence-guided Centroids for Unsupervised Person Re-Identification
Since samples with high confidence are exclusively involved in the formation of centroids, the identity information of low-confidence samples, i. e., boundary samples, are NOT likely to contribute to the corresponding centroid.
Ground Plane Matters: Picking Up Ground Plane Prior in Monocular 3D Object Detection
To pick up the ground plane prior for M3OD, we propose a Ground Plane Enhanced Network (GPENet) which resolves both issues at one go.
Re-parameterizing Your Optimizers rather than Architectures
For the extreme simplicity of model structure, we focus on a VGG-style plain model and showcase that such a simple model trained with a RepOptimizer, which is referred to as RepOpt-VGG, performs on par with or better than the recent well-designed models.
A High-Accuracy Unsupervised Person Re-identification Method Using Auxiliary Information Mined from Datasets
Combined with auxiliary information exploiting modules, our methods achieve mAP of 89. 9% on DukeMTMC, where TOC, STS and SCP all contributed considerable performance improvements.
Ranked #1 on
Unsupervised Person Re-Identification
on MARS
A Free Lunch to Person Re-identification: Learning from Automatically Generated Noisy Tracklets
Extensive experiments on MARS with various manually generated noises show the effectiveness of the proposed framework.
Scaling Up Your Kernels to 31x31: Revisiting Large Kernel Design in CNNs
We revisit large kernel design in modern convolutional neural networks (CNNs).
Ranked #73 on
Image Classification
on ImageNet
TAGPerson: A Target-Aware Generation Pipeline for Person Re-identification
In TAGPerson, we extract information from target scenes and use them to control our parameterized rendering process to generate target-aware synthetic images, which would hold a smaller gap to the real images in the target domain.
RepMLPNet: Hierarchical Vision MLP with Re-parameterized Locality
Our results reveal that 1) Locality Injection is a general methodology for MLP models; 2) RepMLPNet has favorable accuracy-efficiency trade-off compared to the other MLPs; 3) RepMLPNet is the first MLP that seamlessly transfer to Cityscapes semantic segmentation.
Ranked #58 on
Semantic Segmentation
on Cityscapes val
Camera Bias Regularization for Person Re-identification
Due to the discrepancies between cameras caused by illumination, background, or viewpoint, the underlying difficulty for Re-ID is the camera bias problem, which leads to the large gap of within-identity features from different cameras.
LODE: Deep Local Deblurring and A New Benchmark
While recent deep deblurring algorithms have achieved remarkable progress, most existing methods focus on the global deblurring problem, where the image blur mostly arises from severe camera shake.
Emotion Recognition from Multiple Modalities: Fundamentals and Methodologies
In this tutorial, we discuss several key aspects of multi-modal emotion recognition (MER).
Manipulating Identical Filter Redundancy for Efficient Pruning on Deep and Complicated CNN
The existence of redundancy in Convolutional Neural Networks (CNNs) enables us to remove some filters/channels with acceptable performance drops.
Affective Image Content Analysis: Two Decades Review and New Perspectives
Images can convey rich semantics and induce various emotions in viewers.
RepMLP: Re-parameterizing Convolutions into Fully-connected Layers for Image Recognition
We propose RepMLP, a multi-layer-perceptron-style neural network building block for image recognition, which is composed of a series of fully-connected (FC) layers.
Ranked #750 on
Image Classification
on ImageNet
Diverse Branch Block: Building a Convolution as an Inception-like Unit
We propose a universal building block of Convolutional Neural Network (ConvNet) to improve the performance without any inference-time costs.
Computational Emotion Analysis From Images: Recent Advances and Future Directions
Recently, extensive research efforts have been dedicated to understanding the emotions of images.
Automated Model Design and Benchmarking of 3D Deep Learning Models for COVID-19 Detection with Chest CT Scans
The experimental results show that our automatically searched models (CovidNet3D) outperform the baseline human-designed models on the three datasets with tens of times smaller model size and higher accuracy.
RepVGG: Making VGG-style ConvNets Great Again
We present a simple but powerful architecture of convolutional neural network, which has a VGG-like inference-time body composed of nothing but a stack of 3x3 convolution and ReLU, while the training-time model has a multi-branch topology.
Ranked #42 on
Semantic Segmentation
on Cityscapes val
Emotional Semantics-Preserved and Feature-Aligned CycleGAN for Visual Emotion Adaptation
First, we generate an adapted domain to align the source and target domains on the pixel-level by improving CycleGAN with a multi-scale structured cycle-consistency loss.
ResRep: Lossless CNN Pruning via Decoupling Remembering and Forgetting
Via training with regular SGD on the former but a novel update rule with penalty gradients on the latter, we realize structured sparsity.
Shallow Feature Based Dense Attention Network for Crowd Counting
In this paper, we propose a Shallow feature based Dense Attention Network (SDANet) for crowd counting from still images, which diminishes the impact of backgrounds via involving a shallow feature based attention model, and meanwhile, captures multi-scale information via densely connecting hierarchical image features.
Personalized Image Aesthetics Assessment via Meta-Learning With Bilevel Gradient Optimization
Typical image aesthetics assessment (IAA) is modeled for the generic aesthetics perceived by an ``average'' user.
IMRAM: Iterative Matching with Recurrent Attention Memory for Cross-Modal Image-Text Retrieval
Existing methods leverage the attention mechanism to explore such correspondence in a fine-grained manner.
Ranked #18 on
Cross-Modal Retrieval
on Flickr30k
Learning From Multiple Experts: Self-paced Knowledge Distillation for Long-tailed Classification
We refer to these models as 'Experts', and the proposed LFME framework aggregates the knowledge from multiple 'Experts' to learn a unified student model.
Ranked #25 on
Long-tail Learning
on Places-LT
Global Sparse Momentum SGD for Pruning Very Deep Neural Networks
Deep Neural Network (DNN) is powerful but computationally expensive and memory intensive, thus impeding its practical usage on resource-constrained front-end devices.
PDANet: Polarity-consistent Deep Attention Network for Fine-grained Visual Emotion Regression
By optimizing the PCR loss, PDANet can generate a polarity preserved attention map and thus improve the emotion regression performance.
ACNet: Strengthening the Kernel Skeletons for Powerful CNN via Asymmetric Convolution Blocks
We propose Asymmetric Convolution Block (ACB), an architecture-neutral structure as a CNN building block, which uses 1D asymmetric convolutions to strengthen the square convolution kernels.
GRN: Gated Relation Network to Enhance Convolutional Neural Network for Named Entity Recognition
The dominant approaches for named entity recognition (NER) mostly adopt complex recurrent neural networks (RNN), e. g., long-short-term-memory (LSTM).
Ranked #23 on
Named Entity Recognition (NER)
on Ontonotes v5 (English)
Incremental Few-Shot Learning for Pedestrian Attribute Recognition
Pedestrian attribute recognition has received increasing attention due to its important role in video surveillance applications.
Adaptive Region Embedding for Text Classification
Deep learning models such as convolutional neural networks and recurrent networks are widely applied in text classification.
Approximated Oracle Filter Pruning for Destructive CNN Width Optimization
It is not easy to design and run Convolutional Neural Networks (CNNs) due to: 1) finding the optimal number of filters (i. e., the width) at each layer is tricky, given an architecture; and 2) the computational intensity of CNNs impedes the deployment on computationally limited devices.
Centripetal SGD for Pruning Very Deep Convolutional Networks with Complicated Structure
The redundancy is widely recognized in Convolutional Neural Networks (CNNs), which enables to remove unimportant filters from convolutional layers so as to slim the network with acceptable performance drop.
From Zero-shot Learning to Conventional Supervised Classification: Unseen Visual Data Synthesis
Using the proposed Unseen Visual Data Synthesis (UVDS) algorithm, semantic attributes are effectively utilised as an intermediate clue to synthesise unseen visual features at the training stage.
Semantics-Preserving Hashing for Cross-View Retrieval
With benefits of low storage costs and high query speeds, hashing methods are widely researched for efficiently retrieving large-scale data, which commonly contains multiple views, e. g. a news report with images, videos and texts.
Transfer Joint Matching for Unsupervised Domain Adaptation
Visual domain adaptation, which learns an accurate classifier for a new domain using labeled images from an old domain, has shown promising value in computer vision yet still been a challenging problem.
Collective Matrix Factorization Hashing for Multimodal Data
In this paper, we study the problems of learning hash functions in the context of multimodal data for cross-view similarity search.
Image Tag Completion via Image-Specific and Tag-Specific Linear Sparse Reconstructions
Though widely utilized for facilitating image management, user-provided image tags are usually incomplete and insufficient to describe the whole semantic content of corresponding images, resulting in performance degradations in tag-dependent applications and thus necessitating effective tag completion methods.
Transfer Sparse Coding for Robust Image Representation
In this paper, we propose a Transfer Sparse Coding (TSC) approach to construct robust sparse representations for classifying cross-distribution images accurately.
