Object Recognition in Complex Environmental Conditions

• [WACV2018] UG^2: A video benchmark for assessing the impact of image restoration and enhancement on automatic visual recognition [paper] [Dataset]
• [TPAMI2020] Bridging the Gap Between Computational Photography and Visual Recognition [paper]
  • UG2 Challenge CVPR2021 [website]
    • TRACK 1: OBJECT DETECTION IN POOR VISIBILITY ENVIRONMENTS [Website]
      • SUB-CHALLENGE 1.1: OBJECT DETECTION IN THE HAZY CONDITION
      • SUB-CHALLENGE 1.2: FACE DETECTION IN THE LOW-LIGHT CONDITION
• [ICCV2017] AOD-Net: All-in-One Dehazing Network [paper]
• [ECCV2018] Does haze removal help cnn-based image classification? [paper]
• [IJCAI2018] When image denoising meets high-level vision tasks: a deep learning approach [paper] [Code]
• [TIP2020] Connecting Image Denoising and High-Level Vision Tasks via Deep Learning [paper] [Code]
• [TPAMI'2020] DSNet: Joint semantic learning for object detection in inclement weather conditions [paper]
• [AAAI'2022] Image-Adaptive YOLO for Object Detection in Adverse Weather Conditions [paper] [sup] [code]

• [ICCV2011] Blurred target tracking by Blur-driven Tracker [paper]
• [PNAS2016] Atoms of recognition in human and computer vision [paper]
• [QoMEX2016] Understanding how image quality affects deep neural networks [paper]
• [2016] An empirical study on the effects of different types of noise in image classification tasks [paper]
• [ICMLA2017] Google's Cloud Vision API Is Not Robust To Noise [paper]
• [ICCCN2017] A Study and Comparison of Human and Deep Learning Recognition Performance Under Visual Distortions [paper]
• [2017] Examining the Impact of Blur on Recognition by Convolutional Networks [paper]
• [DICTA2017] Enhancing the performance of convolutional neural networks on quality degraded datasets [paper]
• [NIPS2018] Comparing deep neural networks against humans: object recognition when the signal gets weaker [paper]
• [CVIU2019] Getting to Know Low-light Images with The Exclusively Dark Dataset [paper] [Code]
• [ICLR2019] Benchmarking Neural Network Robustness to Common Corruptions and Perturbations [paper] [Code] [Superseded]
• [ICCV2019] Dual Directed Capsule Network for Very Low Resolution Image Recognition [paper]
• [CVPR'2020] Seeing Through FogWithout Seeing Fog: Deep Multimodal Sensor Fusion in Unseen Adverse Weather [paper] [code]
• [2021] RiWNet: A moving object instance segmentation Network being Robust in adverse Weather conditions [paper]

• [CVPR2010] Deconvolutional networks [paper]
• [ICCV2011] Adaptive deconvolutional networks for mid and high level feature learning [paper]
• [VCIP2011] Systematic evaluation of super-resolution using classification [paper]
• [WACV2016] Is image super-resolution helpful for other vision tasks? [paper]
• [ICCV2017] Enhancenet: Single image super-resolution through automated texture synthesis [paper]
• [ICCV2017] AOD-Net: All-in-One Dehazing Network [paper]
• [CVPR2018] DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks [paper] [Code]
• [CVPR2018] Attentive generative adversarial network for raindrop removal from a single image [paper]
• [ECCV2018] The unreasonable effectiveness of texture transfer for single image super-resolution [paper]
• [ECCV2018] Wilddash - creating hazard-aware benchmarks [paper]
• [ECCV2018] Does haze removal help cnn-based image classification? [paper]
• [TPAMI2018] PsyPhy: A Psychophysics Driven Evaluation Framework for Visual Recognition [paper]
• [TIP2019] Benchmarking single-image dehazing and beyond [paper]
  • dataset (Standard and Extended) [google]
  • Belong to TRACK 1: SUB-CHALLENGE 1.1: OBJECT DETECTION IN THE HAZY CONDITION
• [MODA2019] Image Quality and Super Resolution Effects on Object Recognition Using Deep Neural Networks [paper]
• [CVPRW2019] The effects of super-resolution on object detection performance in satellite imagery [paper]
• [CVPR2019] Single image deraining: A comprehensive benchmark analysis [paper] [Code]
  • aim at visual enhancement, not improve recognition
• [2021] Dirty Pixels: Towards End-to-End Image Processing and Perception [paper]
• [TPAMI2022] Exploring Simple and Transferable Recognition-Aware Image Processing [paper] [Code]
• [2023] MoWE: Mixture of Weather Experts for Multiple Adverse Weather Removal [paper]

• [NIPS2013] Adaptive multi-column deep neural networks with application to robust image denoising [paper]
• [TPAMI2015] Adherent raindrop modeling, detection and removal in video [paper]
• [ICIP2016] Joint visual denoising and classification using deep learning [paper]
• [CVPR2018] Classification-driven dynamic image enhancement [paper]
• [IJCAI2018] When image denoising meets high-level vision tasks: a deep learning approach [paper] [Code]
• [AAAI2018] End-to-End United Video Dehazing and Detection [paper]
• [ECCV2018] SOD-MTGAN: Small Object Detection via Multi-Task Generative Adversarial Network [paper]
• [ECCV2020] URIE: Universal Image Enhancement for Visual Recognition in the Wild [paper] [Code]
• [TIP2020] Connecting Image Denoising and High-Level Vision Tasks via Deep Learning [paper] [Code]
• [TPAMI2020] DSNet: Joint semantic learning for object detection in inclement weather conditions [paper]
• [ICONIP2021] Task-Driven Super Resolution: Object Detection in Low-Resolution Images [paper]
• [AAAI2022] Image-Adaptive YOLO for Object Detection in Adverse Weather Conditions [paper] [sup] [code]

• [CVPR2016] Studying Very Low Resolution Recognition Using Deep Networks [paper]
• [ICIP2016] Fine-to-coarse Knowledge Transfer For Low-Res Image Classification [paper]
• [CVPR2018] Domain Adaptive Faster R-CNN for Object Detection in the Wild [paper]
• [ECCV2018] Model Adaptation with Synthetic and Real Data for Semantic Dense Foggy Scene Understanding [paper]
• [ICCV2019] Guided Curriculum Model Adaptation and Uncertainty-Aware Evaluation for Semantic Nighttime Image Segmentation [paper]
• [ECCV2020] Prior-based Domain Adaptive Object Detection for Hazy and Rainy Conditions [paper]
• [ICIP2021] Multiscale Domain Adaptive YOLO for Cross-Domain Object Detection [paper]
• [ACML2021] Domain Adaptive YOLO for One-Stage Cross-Domain Detection [paper]

face recognition (deblurring, super-resolution, hallucination)
person re-identification
Iris recognition 虹膜识别
fingerprint recognition

• [ISSPA2005] Face recognition from super-resolved images [paper]
• [ECBTAS2007] Multi-frame super-resolution for face recognition [paper]
• [ICB2007] Super-Resolved Faces for Improved Face Recognition from Surveillance Video [paper]
• [CVIU2008] Improving long range and high magnification face recognition: Database acquisition, evaluation, and enhancement [paper]
• [CVPR2008] Simultaneous super-resolution and feature extraction for recognition of low-resolution faces [paper]
• [CVPR2009] Facial deblur inference to improve recognition of blurred faces [paper]
• [TNN2010] Super-resolution method for face recognition using nonlinear mappings on coherent features [paper]
• [CVPRW2011] Face recognition in video with closed-loop super-resolution [paper]
• [ICCV2011] Close the loop: Joint blind image restoration and recognition with sparse representation prior [paper]
• [TPAMI2012] A Blur-Robust Descriptor with Applications to Face Recognition [paper]
• [JVCIP2012] Evaluation of image resolution and super-resolution on face recognition performance [paper]
• [CVPR2015] Super-Resolution Person Re-Identification With Semi-Coupled Low-Rank Discriminant Dictionary Learning [paper]
• [BMVC2015] Convolutional neural networks for direct text deblurring [paper]
• [SIU2016] Facial image super resolution using sparse representation for improving face recognition in surveillance monitoring [paper]
• [AMDO2016] Convolutional Neural Network Super Resolution for Face Recognition in Surveillance Monitoring [paper]
• [ECCV2016] Learning High-Order Filters for Efficient Blind Deconvolution of Document Photographs [paper]
• [ECCV2016] Colorful Image Colorization [paper]
• [2016] Deep joint face hallucination and recognition [paper]
• [BIOSIG2016] How image degradations affect deep cnn-based face recognition? [paper]
• [ICIP2017] Quality-adaptive deep learning for pedestrian detection [paper]
• [ICCV2017] VPGNet: Vanishing Point Guided Network for Lane and Road Marking Detection and Recognition [paper]
• [ICCV2019] Delving into Robust Object Detection from Unmanned Aerial Vehicles: A Deep Nuisance Disentanglement Approach [paper]
• [ICIP2019] Quality-adaptive deep learning for pedestrian detection [paper]