Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks116 from public data
- Deep Residual Learning for Image Recognition241,838
This work presents a residual learning framework to ease the training of networks that are substantially deeper than those used previously, and provides comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth.
- Identity Mappings in Deep Residual Networks11,460
The propagation formulations behind the residual building blocks suggest that the forward and backward signals can be directly propagated from one block to any other block, when using identity mappings as the skip connections and after-addition activation.
- ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices9,446
An extremely computation-efficient CNN architecture named ShuffleNet is introduced, which is designed specially for mobile devices with very limited computing power (e.g., 10-150 MFLOPs), to greatly reduce computation cost while maintaining accuracy.
- ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design6,868
This work proposes to evaluate the direct metric on the target platform, beyond only considering FLOPs, and derives several practical guidelines for efficient network design, called ShuffleNet V2.
- 2,946
- Unified Perceptual Parsing for Scene Understanding2,690
A multi-task framework called UPerNet and a training strategy are developed to learn from heterogeneous image annotations and it is shown that it is able to effectively segment a wide range of concepts from images.
- RepVGG: Making VGG-style ConvNets Great Again2,660
A simple but powerful architecture of convolutional neural network, which has a VGG-like inference-time body composed of nothing but a stack of 3 × 3 convolution and ReLU, while the training-time model has a multi-branch topology.
- 1,360
- 1,124
- Single Path One-Shot Neural Architecture Search with Uniform Sampling1,060
A Single Path One-Shot model is proposed to construct a simplified supernet, where all architectures are single paths so that weight co-adaption problem is alleviated.
- You Only Look One-level Feature788
This paper revisits feature pyramids networks (FPN) for one-stage detectors and points out that the success of FPN is due to its divide-and-conquer solution to the optimization problem in object detection rather than multi-scale feature fusion.
- DFANet: Deep Feature Aggregation for Real-Time Semantic Segmentation716
This paper introduces an extremely efficient CNN architecture named DFANet for semantic segmentation under resource constraints that substantially reduces the number of parameters, but still obtains sufficient receptive field and enhances the model learning ability, which strikes a balance between the speed and segmentation performance.
- MetaPruning: Meta Learning for Automatic Neural Network Channel Pruning650
A novel meta learning approach for automatic channel pruning of very deep neural networks by training a PruningNet, a kind of meta network, which is able to generate weight parameters for any pruned structure given the target network.
- Anchor DETR: Query Design for Transformer-Based Detector631
The proposed detector, called Anchor DETR, can achieve better performance and run faster than the DETR with 10x fewer training epochs, and an attention variant, which can reduce the memory cost while achieving similar or better performance than the standard attention in DETR.
- Repulsion Loss: Detecting Pedestrians in a Crowd591
This paper first explores how a state-of-the-art pedestrian detector is harmed by crowd occlusion via experimentation, and proposes a novel bounding box regression loss specifically designed for crowd scenes, termed repulsion loss.
- ExFuse: Enhancing Feature Fusion for Semantic Segmentation572
It is found that introducing semantic information into low-level features and high-resolution details into high-level features is more effective for the later fusion, and a new framework, named ExFuse, is proposed to bridge the gap between low-level and high-level features.
- OTA: Optimal Transport Assignment for Object Detection550
This paper innovatively revisit the label assignment from a global perspective and proposes to formulate the assigning procedure as an Optimal Transport (OT) problem – a well-studied topic in Optimization Theory.
- Meta-SR: A Magnification-Arbitrary Network for Super-Resolution539
This work proposes a novel method called Meta-SR to firstly solvesuper-resolution of arbitrary scale factor (including non-integer scale factors) with a single model and shows the superi-ority of the Meta-Upscale.
- Deep Learning with Low Precision by Half-Wave Gaussian Quantization537
An half-wave Gaussian quantizer (HWGQ) is proposed for forward approximation and shown to have efficient implementation, by exploiting the statistics of of network activations and batch normalization operations, and to achieve much closer performance to full precision networks than previously available low-precision networks.
- High-Order Information Matters: Learning Relation and Topology for Occluded Person Re-Identification512
This paper proposes a novel framework by learning high-order relation and topology information for discriminative features and robust alignment for occluded person re-identification and significantly outperforms state-of-the-art mAP scores on Occluded-Duke dataset.
- 462
- 454
- Perceive Where to Focus: Learning Visibility-Aware Part-Level Features for Partial Person Re-Identification388
A Visibility-aware Part Model (VPM) is proposed for partial re-ID, which learns to perceive the visibility of regions through self-supervision, which significantly improves the learned feature representation and the achieved accuracy is on par with the state of the art.
- FFB6D: A Full Flow Bidirectional Fusion Network for 6D Pose Estimation376
This work proposes FFB6D, a Full Flow Bidirectional fusion network designed for 6D pose estimation from a single RGBD image, which learns to combine appearance and geometry information for representation learning as well as output representation selection.
- Distribution Alignment: A Unified Framework for Long-tail Visual Recognition376
This work develops an adaptive calibration function that enables us to adjust the classification scores for each data point in the two-stage learning to balance the class prior, which provides a flexible and unified solution to diverse scenarios in visual recognition tasks.
- DetNet: Design Backbone for Object Detection337
DetNet is proposed, which is a novel backbone network specifically designed for object detection that includes the extra stages against traditional backbone network for image classification, while maintains high spatial resolution in deeper layers.
- ThunderNet: Towards Real-Time Generic Object Detection on Mobile Devices292
benefit from the highly efficient backbone and detection part design, ThunderNet surpasses previous lightweight one-stage detectors with only 40% of the computational cost on PASCAL VOC and COCO benchmarks.
- End-to-End Human Object Interaction Detection with HOI Transformer274
Hoi Transformer streamlines the HOI pipeline by eliminating the need for many hand-designed components and directly predicts HOI instances in parallel, surpassing previous methods with the advantage of being much simpler.
- Learning Human-Object Interaction Detection Using Interaction Points256
This paper proposes a novel fully-convolutional approach that directly detects the interactions between human-object pairs and predicts interaction points, which directly localize and classify the inter-action.
- End-to-End Object Detection with Fully Convolutional Network247
A Prediction-aware One-To-One (POTO) label assignment for classification is introduced to enable end-to-end detection, which obtains comparable performance with NMS and a simple 3D Max Filtering is proposed to utilize the multi-scale features and improve the discriminability of convolutions in the local region.
- Detection in Crowded Scenes: One Proposal, Multiple Predictions231
The key of the approach is to let each proposal predict a set of correlated instances rather than a single one in previous proposal-based frameworks, which can effectively handle the difficulty of detecting highly overlapped objects.
- Generalized Few-Shot Object Detection without Forgetting224
Extensive experiments show that Retentive R-CNN significantly outperforms state-of-the-art methods on overall performance among all settings as it can achieve competitive results on few-shot classes and does not degrade the base class performance at all.
- Content-Aware Unsupervised Deep Homography Estimation204
This work proposes an unsupervised deep homography method with a new architecture design that outperforms the state-of-the-art including deep solutions and feature-based solutions.
- Fully Convolutional Networks for Panoptic Segmentation202
This approach aims to represent and predict foreground things and background stuff in a unified fully convolutional pipeline and outperforms previous box-based and -free models with high efficiency on COCO, Cityscapes, and Mapillary Vistas datasets with single scale input.
- 181
- 179
- Dynamic Region-Aware Convolution160
DRConv is an effective and elegant method for handling complex and variable spatial information distribution that can substitute standard convolution in any existing networks for its plug-and-play property, especially to power convolution layers in efficient networks.
- BorderDet: Border Feature for Dense Object Detection154
This paper designs a novel detection architecture called BorderDet, which explicitly exploits the border information for stronger classification and more accurate localization and proposes a simple and efficient operator called Border-Align to extract "border features" from the extreme point of the border to enhance the point feature.
- WeightNet: Revisiting the Design Space of Weight Networks129
The WeightNet, composed entirely of (grouped) fully-connected layers, is used to directly output the convolutional weight, and outperforms existing approaches on both ImageNet and COCO detection tasks, achieving better Accuracy-FLOPs and Accuracy-Parameter trade-offs.
- ADNet: Attention-guided Deformable Convolutional Network for High Dynamic Range Imaging116
This paper presents an attention-guided deformable convolutional network for hand-held multi-frame high dynamic range (HDR) imaging, namely ADNet, and proposes to align the gamma-corrected images in the feature-level with a Pyramid, Cascading and Deformable alignment module.
- Funnel Activation for Visual Recognition112
A conceptually simple but effective funnel activation for image recognition tasks, called Funnel activation (FReLU), that extends ReLU and PReLU to a 2D activation by adding a negligible overhead of spatial condition.
- Points as Queries: Weakly Semi-supervised Object Detection by Points110
A new detector is introduced, Point DETR, which extends DETR by adding a point encoder, which achieves a promising performance when using 20% fully labeled data from COCO, and the point annotations bring over 10 points in various AR metrics.
- Tree Energy Loss: Towards Sparsely Annotated Semantic Segmentation109
A novel tree energy loss for SASS is proposed by providing semantic guidance for unlabeled pixels by sequentially applying these affinities to the net-work prediction, achieving dynamic online self-training.
- UPFlow: Upsampling Pyramid for Unsupervised Optical Flow Learning96
This work designs a self-guided upsample module to tackle the interpolation blur problem caused by bilinear upsampling between pyramid levels, and proposes a pyramid distillation loss to add supervision for intermediate levels via distilling the finest flow as pseudo labels.
- DetNAS: Neural Architecture Search on Object Detection.94
This paper proposes DetNAS to automatically search neural architectures for the backbones of object detectors, formulated into a supernet and the search method relies on evolution algorithm (EA).
- Individual identification of dairy cows based on convolutional neural networks94
A novel contactless cow identification method based on convolutional neural networks is proposed that outperformed existing experiments in cow identification and the potential for the method to be applied to other livestock.
- 83
- Stitcher: Feedback-driven Data Provider for Object Detection76
Stitcher is presented, a feedback-driven data provider, which aims to train object detectors in a balanced way, and steadily improves performance by a large margin in all settings, especially for small objects, with nearly no additional computation in both training and testing stages.
- Real-time Object Detection for Streaming Perception71
This paper builds a simple and effective frame-work for streaming perception that achieves competitive performance on Argoverse-HD dataset and improves the AP by 4.9% compared to the strong baseline, validating its effectiveness.
- Angle-Based Search Space Shrinking for Neural Architecture Search70
Comp comprehensive evidences are provided showing that, in weight-sharing supernet, the proposed metric is more stable and accurate than accuracy-based and magnitude-based metrics to predict the capability of child models.
- Neural Architecture Search with Random Labels69
A novel NAS framework based on ease-of-convergence hypothesis, which requires only random labels during searching, which achieves comparable or even better results compared with state-of theart NAS methods such as PC-DARTS, Single Path One-Shot, even though the counterparts utilize full ground truth labels for searching.
- IQDet: Instance-wise Quality Distribution Sampling for Object Detection66
A dense object detector with an instance-wise sampling strategy, named IQDet, which automatically selects training samples in a probabilistic manner and trains with more high-quality samples, and is completely cost-free in inference time.
- 66
- Fully Convolutional Networks for Panoptic Segmentation with Point-based Supervision65
A conceptually simple, strong, and efficient framework for fully- and weakly-supervised panoptic segmentation, called Panoptic FCN, which encodes each object instance or stuff category with the proposed kernel generator and produces the prediction by convolving the high-resolution feature directly.
- BSRT: Improving Burst Super-Resolution with Swin Transformer and Flow-Guided Deformable Alignment61
This work combines optical flows and deformable convolutions, hence the BSRT can handle misalignment and aggregate the potential texture information in multi-frames more efficiently and wins the championship in the NTIRE2022 Burst Super-Resolution Challenge.
- 49
- Sparse-Coded Features for Image Retrieval45
This paper illustrates that Fisher Vector, VLAD and BOF can be uniformly derived in two steps: i Encoding – separately map each local descriptor into a code, and ii Pooling – aggregate all codes from one image into a single vector.
- LGD: Label-Guided Self-Distillation for Object Detection43
Compared with a classical teacher-based method FGFI, LGD not only performs better without requiring pretrained teacher but also reduces 51% training cost beyond inherent student learning.
- Attentive Normalization for Conditional Image Generation39
This paper characterize long-range dependence with attentive normalization (AN), which is an extension to traditional instance normalization, which enhances consistency between distant regions with semantic correspondence.
- 38
- GridFace: Face Rectification via Learning Local Homography Transformations37
This paper proposes a method, called GridFace, to reduce facial geometric variations and improve the recognition performance, which rectifies the face by local homography transformations, which are estimated by a face rectification network.
- Rendering Portraitures from Monocular Camera and Beyond30
This work introduces an automatic system that achieves portrait DoF rendering for monocular cameras using Convolutional Neural Networks and trains a spatially-variant Recursive Neural Network to learn and accelerate this rendering process.
- LabelEnc: A New Intermediate Supervision Method for Object Detection29
A novel label encoding function is introduced, mapping the ground-truth labels into latent embedding, acting as an auxiliary intermediate supervision to the detection backbone during training, approximating the "desired" intermediate representations for the target object detector.
- 27
- 25
- 25
- Differentiable Architecture Search with Random Features23
This paper investigates the expres-sive power of the supernet in DARTS and derives a new setup of DARTS paradigm with only training Batch-Norm and theoretically finds that random features dilute the auxiliary connection role of skip-connection in supernet optimization and enable search algorithm focus on fairer operation selection, thereby solving the performance collapse problem.
- Joint Multi-Dimension Pruning via Numerical Gradient Update22
Joint multi-dimension pruning discovers a better status than previous studies that focused on individual dimensions solely, as the method is optimized collaboratively across the three dimensions in a single end-to-end training and it is more efficient than the previous exhaustive methods.
- 19
- 19
- Data-driven control against false data injection attacks18
A general FDI attack model is presented, which imposes minimally constraints on the switching frequency of attack channels and the magnitude of attack matrices, and a dynamic state feedback control law is designed based on offline and online input-state data, which adapts to the channel switching of FDI attacks.
- 16
- Scale Enhancement Pyramid Network for Small Object Detection from UAV Images14
A simple yet effective scale enhancement pyramid network (SEPNet) that combines multi-scale atrous convolution and multi-branch grouped convolution to model global relationships, effectively alleviating information conflict caused by the fusion of adjacent features.
- Lightweight UAV Object-Detection Method Based on Efficient Multidimensional Global Feature Adaptive Fusion and Knowledge Distillation13
The MGFAFNET is proposed, an innovative detection method for UAV platforms that demonstrates remarkable improvements over state-of-the-art object-detection methods, as demonstrated by rigorous evaluations conducted using the VisDrone2021 benchmark and a meticulously assembled proprietary dataset.
- DS-UI: Dual-Supervised Mixture of Gaussian Mixture Models for Uncertainty Inference in Image Recognition13
Experimental results show the DS-UI outperforms the state-of-the-art UI methods in misclassification detection and finds statistically significant improvements in open-set out- of-domain/-distribution detection.
- 13
- 10
- 10
- 9
- When NAS Meets Trees: An Efficient Algorithm for Neural Architecture Search9
A new NAS method called TNAS (NAS with trees), which improves search efficiency by exploring only a small number of architectures while also achieving a higher search accuracy, is proposed.
- Neural Diffusion Distance for Image Segmentation9
With the learned diffusion distance, a hierarchical image segmentation method outperforming previous segmentation methods is proposed and achieved promising results on PASCAL VOC 2012 segmentation dataset.
- Identifying Mouse Autoimmune Uveitis from Fundus Photographs Using Deep Learning7
The proposed deep learning model is able to identify three severity levels of EAU with high accuracy and achieved high accuracy on independent validation subsets, reflecting a substantial degree of generalizability.
- Joint Multi-Dimension Pruning7
Joint multi-dimension pruning is presented, a new perspective of pruning a network on three crucial aspects: spatial, depth and channel simultaneously, as this method is optimized collaboratively across the three dimensions in a single end-to-end training.
- Data-Driven Dynamic Output Feedback Nash Strategy for Multi-Player Non-Zero-Sum Games6
This paper investigates the multi-player non-zero-sum game problem for unknown linear continuous-time systems with unmeasurable states and proposes a data-driven learning control approach which can form Nash equilibrium solution to the multi-player non-zero-sum game problem.
- Data-driven optimal cooperative tracking control for heterogeneous multi-agent systems6
This paper presents a novel hierarchical control scheme for solving the data-driven optimal cooperative tracking control problem of heterogeneous multi-agent systems where followers cannot communicate with the leader and a prescribed-time fully distributed observer is devised to estimate the leader's state for each follower.
- 6
- A Wordnet-Based Approach to Feature Selection in Text Categorization6
In this method, word tendency, which takes related words into consideration, is used to select best terms and the best performance is usually achieved in fewer features.
- Automatic Text Categorization Based on K-Nearest Neighbor6
A method that integrates language information and statistical information from the training corpus is put forward and shows that the method improves the accuracy of the categorization.
- Diffusion-based arbitrary-scale magnetic resonance image super-resolution via progressive k-space reconstruction and denoising5
A unified diffusion-based framework for arbitrary-scale in-plane MR image SR, dubbed Progressive Reconstruction and Denoising Diffusion Model (PRDDiff), which demonstrates that PRDDiff outperforms previous MR image super-resolution methods in term of reconstruction accuracy, generalization, and downstream lesion segmentation accuracy and CP classification performance.
- 4
- 4
- 4
- Automatic detection of mortar loss on masonry building facades based on deep learning3
A two-step automatic detection method for mortar loss, combining deep learning models with sliding window algorithm, was proposed and Experimental results demonstrate its capability to accurately and efficiently detect mortar loss in practical scenarios, with strong generalization performance and noise resistance.
- Multi-modal 6-DoF object pose tracking: integrating spatial cues with monocular RGB imagery3
This work introduces a novel multi-modal 6-DoF pose estimation framework that uses RGB images as the primary input and integrates spatial cues, including keypoint heatmaps and affinity fields, through a spatially aligned approach inspired by the Trans-UNet architecture.
- 3
- A localization method of manipulator towards achieving more precision control3
This article researched the robot arm positioning method via monocular vision by building a vision system model and designed the style of a cooperative target for target positioning, and combining the principle of pose estimation on the PNP (Perspective‐n‐Point) problem with the results of the visual system calibration to realize the positioning of the target.
- Stacking Ensemble Learning in Deep Domain Adaptation for Ophthalmic Image Classification3
This work proposes a novel method, named SELDA, for stacking ensemble learning via extending three domain adaptation methods for effectively solving real-world problems by stacking the output predictions of these three well-performing domain adaptation models.
- 3
- 2
- A new training algorithm for Quantum Neural Networks2
Using the theory of Constrained Optimization, a new training algorithm is proposed in this paper by introducing penalty function into the steepest descent method and results indicate that the new algorithm can increase the training speed and improve the performance of the trained network remarkably.
- The Application of GPS in Area Measurement2
The function of AT89C51 its peripheral extended chip in receiving and send ing GPS data, and the flow chat of the software are presented and the curve fiting is based on the Least Square principle.
- 1
- 1
- 1
- A Novel Structure Design and Training Algorithm for Quantum Neural Network1
The QNN with multi-tan-sigmoid transfer function and a novel training algorithm which combines the two objective functions are proposed and Experimental results demonstrate the effectiveness of the structure improvement and the new training algorithm.
- Estimating the Generalization Performance of Polynomial SVM Classifier for Text Categorization1
The research indicates over-fitting problems occur as the polynomial order increases, and SVM's generalization performance decreases drastically if too many features are used, so feature selection is necessary.
- –
- An Intelligent Multi-Agent Framework for Automated Drafting and Compliance Optimization of Cadre Evaluation Materials–
Experimental analysis demonstrates that the proposed system significantly reduces manual workload while improving coherence, compliance, and efficiency, thus providing a scalable and intelligent solution for government and organizational evaluation tasks.
- HTS-LB: Hypergraph tree search for learning branch–
A hypergraph tree search framework for learning branch (HTS-LB) is proposed that is effective and outperforms popular machine learning algorithms in terms of branching accuracy, branch and bound nodes, and the dual-primal gap.
- –
- Dynamic analysis of railroad track faults based on neural network algorithms–
The model in this paper based on a neural network algorithm can achieve efficient and accurate diagnosis of railroad track faults and shows that the model can achieve efficient and accurate diagnosis of railroad track faults.
- –
- –
- –
- Auto-exposure Algorithm for CCD Camera Using Image Histogram–
The experimental results show that images obtained by the proposed algorithm present appropriate brightness, proving that the algorithm not only has a good capability to adapt to various scenes, but also it can meet the requirement of automatic exposure of CCD cameras.
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-14. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.