논문 리뷰10

• 

  [GAN] High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs (CVPR 2018) High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs We present a new method for synthesizing high-resolution photo-realistic images from semantic label maps using conditional generative adversarial networks (conditional GANs). Conditional GANs have enabled a variety of applications, but the results are ofte arxiv.org Abstract 이 논문에선 conditional generative adversarial .. 논문 리뷰 2022. 4. 6.

• 

  [이상 탐지] Pixel-wise Anomaly Detection in Complex Driving Scenes (CVPR 2021) Pixel-wise Anomaly Detection in Complex Driving Scenes The inability of state-of-the-art semantic segmentation methods to detect anomaly instances hinders them from being deployed in safety-critical and complex applications, such as autonomous driving. Recent approaches have focused on either leveraging segmen arxiv.org Abstract State-of-the-art semantic segmenatation은 이상 객체 (anomlay instances)를.. 논문 리뷰 2022. 3. 15.

• 

  [이상 탐지] CutPaste: Self-Supervised Learning for Anomaly Detection and Localization (CVPR 2021) CutPaste: Self-Supervised Learning for Anomaly Detection and Localization We aim at constructing a high performance model for defect detection that detects unknown anomalous patterns of an image without anomalous data. To this end, we propose a two-stage framework for building anomaly detectors using normal training data only. W arxiv.org Abstract Anomalous data 없이 unknown anomalous pattern을 이미지.. 논문 리뷰 2022. 2. 24.

• 

  [ResNet] Deep Residual Learning for Image Recognition Deep Residual Learning for Image Recognition Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with arxiv.org Abstract 신경망이 깊을수록 training이 어려워진다. Residual Network는 심층적인 네트워크의 training을 용이하게 하기 위해 .. 논문 리뷰 2022. 2. 15.

• 

  [HRNet] Deep High-Resolution Representation Learning for Visual Recognition 논문 리뷰 Deep High-Resolution Representation Learning for Visual Recognition High-resolution representations are essential for position-sensitive vision problems, such as human pose estimation, semantic segmentation, and object detection. Existing state-of-the-art frameworks first encode the input image as a low-resolution represen arxiv.org 논문 리뷰 2022. 2. 14.

• 

  YOLO (You Only Look Once) v4 Review 1개의 GPU만으로 BoF, BoS 기법을 적용하여 효율적이고 정확한 Object Detection이 가능 기존 YOLO v1 ~ v3에서 바뀐 점? 기존 YOLO는 작은 물체에 취약 YOLO v4는 다양한 작은 object들을 잘 검출하기 위해 input resolution을 크게 사용 (224, 256 → 512) Receptive field를 물리적으로 키워주기 위해 layer 수를 늘림 하나의 image에서 다양한 종류, 크기의 object들을 검출하기 위해 parameter 수를 키움 속도 관점에서 이득을 보기 위해 CSPNet 기반의 backbone을 사용 YOLO v4의 구조 Backbone : CSP-Darknet 53 Neck : SPP (Spatial Pyramid Pooling), P.. 논문 리뷰 2021. 10. 18.

• 

  [차선 인식] PINet : Key Points Estimation and Point Instance Segmentation Approach for Lane Detection GitHub - koyeongmin/PINet_new Contribute to koyeongmin/PINet_new development by creating an account on GitHub. github.com GitHub - koyeongmin/PINet Contribute to koyeongmin/PINet development by creating an account on GitHub. github.com Key Points Estimation and Point Instance Segmentation Approach for Lane Detection Perception techniques for autonomous driving should be adaptive to various envir.. 논문 리뷰 2021. 8. 10.

• 

  [얼굴 인식] MagFace: A Universal Representation for Face Recognition and Quality Assessment GitHub - IrvingMeng/MagFace: MagFace: A Universal Representation for Face Recognition and Quality Assessment, CVPR2021, Oral MagFace: A Universal Representation for Face Recognition and Quality Assessment, CVPR2021, Oral - GitHub - IrvingMeng/MagFace: MagFace: A Universal Representation for Face Recognition and Quality A... github.com MagFace: A Universal Representation for Face Recognition and .. 논문 리뷰 2021. 8. 4.

• 

  [얼굴 인식] ArcFace: Additive Angular Margin Loss for Deep Face Recognition ArcFace: Additive Angular Margin Loss for Deep Face Recognition One of the main challenges in feature learning using Deep Convolutional Neural Networks (DCNNs) for large-scale face recognition is the design of appropriate loss functions that enhance discriminative power. Centre loss penalises the distance between the d arxiv.org GitHub - deepinsight/insightface: State-of-the-art 2D and 3D Face A.. 논문 리뷰 2021. 8. 4.

• 

  [얼굴 인식] SpereFace : Deep Hypershpere Embedding for Face Recognition 논문 리뷰 SphereFace: Deep Hypersphere Embedding for Face Recognition This paper addresses deep face recognition (FR) problem under open-set protocol, where ideal face features are expected to have smaller maximal intra-class distance than minimal inter-class distance under a suitably chosen metric space. However, few existi arxiv.org GitHub - wy1iu/sphereface: Implementation for in CVPR'17. Implementatio.. 논문 리뷰 2021. 7. 22.