论文汇总

汇总本站发布过的全部论文调研，共 106 篇，按领域与年份整理。覆盖多模态、模型训练、图像分类、目标检测/分割、图像检索、OCR、人脸检测/识别、模型蒸馏/剪枝、视频识别/压缩等 13 个方向。

多模态（3篇）

• 2022
  • Chinese CLIP: Contrastive Vision-Language Pretraining in Chinese
  • Reproducible scaling laws for contrastive language-image learning
• 2021
  • Learning Transferable Visual Models From Natural Language Supervision

模型训练（12篇）

• Pytorch
  • How to Train State-Of-The-Art Models Using TorchVision’s Latest Primitives
  • TorchVision Object Detection Finetuning Tutorial
  • Transfer Learning for Computer Vision Tutorial
  • VISUALIZING MODELS, DATA, AND TRAINING WITH TENSORBOARD
• 2019
  • RandAugment: Practical automated data augmentation with a reduced search space
  • Fast AutoAugment
• 2018
  • AutoAugment
  • A DISCIPLINED APPROACH TO NEURAL NETWORK HYPER-PARAMETERS: PART 1 – LEARNING RATE, BATCH SIZE, MOMENTUM, AND WEIGHT DECAY
• 2017
  • Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour
• 2016
  • SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS
• 2015
  • Cyclical Learning Rates for Training Neural Networks
• 2014
  • Microsoft COCO: Common Objects in Context

图像分类（32篇）

• 2022
  • A ConvNet for the 2020s
• 2021
  • EfficientNetV2: Smaller Models and Faster Training
  • RepVGG: Making VGG-style ConvNets Great Again
• 2020
  • Higher accuracy on vision models with EfficientNet-Lite
  • Model Rubik’s Cube: Twisting Resolution, Depth and Width for TinyNets
  • ResNeSt: Split-Attention Networks
• 2019
  • GhostNet: More Features from Cheap Operations
  • ACNet: Strengthening the Kernel Skeletons for Powerful CNN via Asymmetric Convolution Blocks
  • EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
  • Searching for MobileNetV3
  • Selective Kernel Networks
• 2018
  • MnasNet: Platform-Aware Neural Architecture Search for Mobile
  • ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design
  • MobileNetV2: Inverted Residuals and Linear Bottlenecks
• 2017
  • Squeeze-and-Excitation Networks
  • ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile
  • MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• 2016
  • Aggregated Residual Transformations for Deep Neural Networks
  • Identity Mappings in Deep Residual Networks
  • Densely Connected Convolutional Networks
  • SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size
  • Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning
• 2015
  • Rethinking the Inception Architecture for Computer Vision
  • [GoogLeNet]Inception_v2
  • [GoogLeNet]Inception_v3
  • Deep Residual Learning for Image Recognition
  • Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• 2014
  • Going deeper with convolutions
  • very deep convolutional networks for large-scale image recognition
• 2013
  • Network in Network
  • Visualizing and Understanding Convolutional Networks
• 2012
  • ImageNet Classification with Deep Convolutional Neural Networks

目标检测/分割（13篇）

• 2019
  • YOLACT: Real-time Instance Segmentation
  • Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regression
• 2018
  • YOLOv3: An Incremental Improvement
• 2017
  • Focal Loss for Dense Object Detection
• 2016
  • Feature Pyramid Networks for Object Detection
  • YOLO9000: Better, Faster, Stronger
• 2015
  • Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
  • SSD: Single Shot MultiBox Detector
  • You Only Look Once: Unified, Real-Time Object Detection
  • Fast R-CNN
• 2014
  • Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
• 2013
  • Selective Search for Object Recognition
  • Rich feature hierarchies for accurate object detection and semantic segmentation

图像检索（19篇）

• 2020
  • 3rd Place Solution to Google Landmark Retrieval 2020
  • PyRetri: A PyTorch-based Library for Unsupervised Image Retrieval by Deep Convolutional Neural Networks
• 2019
  • A Benchmark on Tricks for Large-scale Image Retrieval
  • Combination of Multiple Global Descriptors for Image Retrieval
  • MultiGrain: a unified image embedding for classes and instances
  • 2nd Place and 2nd Place Solution to Kaggle Landmark Recognition and Retrieval Competition 2019
  • Learning with Average Precision: Training Image Retrieval with a Listwise Loss
  • Multi-Similarity Loss with General Pair Weighting for Deep Metric Learning
  • Revisiting Oxford and Paris: Large-Scale Image Retrieval Benchmarking
• 2017
  • Fine-tuning CNN Image Retrieval with No Human Annotation
• 2016
  • End-to-end Learning of Deep Visual Representations for Image Retrieval
  • CNN Image Retrieval Learns from BoW: Unsupervised Fine-Tuning with Hard Examples
  • Deep Image Retrieval: Learning global representations for image search
• 2015
  • Particular object retrieval with integral max-pooling of CNN activations
  • Cross-dimensional Weighting for Aggregated Deep Convolutional Features
  • Deep Learning of Binary Hash Codes for Fast Image Retrieval
  • Aggregating Deep Convolutional Features for Image Retrieval
• 2014
  • Deep Learning for Content-Based Image Retrieval: A Comprehensive Study
  • Visual Instance Retrieval with Deep Convolutional Networks

OCR（3篇）

• 2024
  • Towards End-to-End License Plate Detection and Recognition: A Large Dataset and Baseline
• 2018
  • LPRNet: License Plate Recognition via Deep Neural Networks
• 2015
  • An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition

人脸检测/识别（3篇）

• 2024
  • WIDER FACE: A Face Detection Benchmark
• 2021
  • YOLO5Face: Why Reinventing a Face Detector
• 2015
  • FaceNet: A Unified Embedding for Face Recognition and Clustering

模型蒸馏（6篇）

• 2021
  • Distilling Knowledge via Knowledge Review
• 2019
  • A Comprehensive Overhaul of Feature Distillation
• 2017
  • A Gift from Knowledge Distillation: Fast Optimization, Network Minimization and Transfer Learning
• 2016
  • Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfer
• 2014
  • FitNets: Hints for Thin Deep Nets
  • Distilling the Knowledge in a Neural Network

模型剪枝（6篇）

• 2020
  • Lossless CNN Channel Pruning via Decoupling Remembering and Forgetting
• 2018
  • Rethinking the Value of Network Pruning
• 2017
  • Learning Efficient Convolutional Networks through Network Slimming
• 2016
  • Learning Structured Sparsity in Deep Neural Networks
• 2015
  • Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding
  • Learning both Weights and Connections for Efficient Neural Networks

视频识别（4篇）

• 2018
  • TSM: Temporal Shift Module for Efficient Video Understanding
• 2017
  • Temporal Relational Reasoning in Videos
• 2016
  • Temporal Segment Network
• 2014
  • Two-Stream Convolutional Networks for Action Recognition in Videos

细粒度分类（1篇）

• 2019
  • Destruction and Construction Learning for Fine-grained Image Recognition

行人属性识别（2篇）

• 2023
  • RethinkingPAR-行人属性识别
• 2021
  • Rethinking of Pedestrian Attribute Recognition: A Reliable Evaluation under Zero-Shot Pedestrian Identity Setting

视频压缩（2篇）

• 2020
  • OpenDVC: An Open Source Implementation of the DVC Video Compression Method
• 2018
  • DVC: An End-to-end Deep Video Compression Framework

相关阅读

• 图像分类
• 基于内容的图像检索
• 度量学习
• 知识迁移
• 网络剪枝
• 视频压缩
• 视频分类
• 细粒度分类
• 目标检测训练框架