cb

Liu, Zuhao, et al. "Generating anomalies for video anomaly detection with prompt-based feature mapping." Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2023.https://openaccess.thecvf.com/content/CVPR2023/papers/Liu_Generating_Anomalies_for_Video_Anomaly_Detection_With_Prompt-Based_Feature_Mapping_CVPR_2023_paper.pdf 0. Abstract감시 영상에서의 이상 탐지는 훈련 과정에서 normal vi..

Chen, Weiling, et al. "TEVAD: Improved video anomaly detection with captions." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2023.openaccess.thecvf.comhttps://github.com/coranholmes/TEVAD GitHub - coranholmes/TEVAD: Official implementation for paper TEVAD: Improved video anomaly detection with captionsOfficial implementation for paper TEVAD: Improved video an..

Self-Supervised Learning from Images with a Joint-Embedding Predictive ArchitectureMahmoud Assran, Quentin Duval, Ishan Misra, Piotr Bojanowski, Pascal Vincent, Michael Rabbat, Yann LeCun, Nicolas Ballashttps://arxiv.org/pdf/2301.08243 Abstract본 논문에서는 hand-crafted data-augmentations에 의존하지 않고 highly semantic image representations을 학습하는 방법, I-JEPA(Image-based Joint-Embedding Predictive Architectur..

One-2-3-45: Any Single Image to 3D Mesh in 45 Seconds without Per-Shape OptimizationMinghua Liu, Chao Xu, Haian Jin, Linghao Chen, Mukund Varma T, Zexiang Xu, Hao Suhttps://arxiv.org/abs/2306.16928 One-2-3-45: Any Single Image to 3D Mesh in 45 Seconds without Per-Shape OptimizationSingle image 3D reconstruction is an important but challenging task that requires extensive knowledge of our natural..

Pix2Pix: Image-to-Image Translation with Conditional Adversarial NetworksPhillip Isola, Jun-Yan Zhu, Tinghui Zhou, Alexei A. Efroshttps://arxiv.org/pdf/1611.07004 0. Abstract조건부 적대적 신경망(conditional adversarial networks)image-to-image tranlation problemsinput 이미지에서부터 output 이미지로부터 매핑하는 방법을 배움reconstructing objects, colorizing images 등등을 가능하게 함이는 별도의 매개변수 조정(parameter tweaking) 없이도 쉽게 적용이 가능할 것을 보..

Voxelnet: End-to-end learning for point cloud based 3d object detectionZhou, Yin, and Oncel Tuzel. "Voxelnet: End-to-end learning for point cloud based 3d object detection - CVPR 2018 https://arxiv.org/pdf/1711.06396 0. Abstract자율 주행 네비게이션, VR 등에 적용되기 위해 3D point clouds은 정확한 detection을 해내야 함 이전 연구들에서는 sparse한 LiDAR point cloud를 RPN으로 다루기 위해 bird-eye view와 같은 hand-crafted feature extractions를 사용함..

Frustum pointnets for 3d object detection from rgb-d data.https://arxiv.org/pdf/1711.08488 AbstractRGBD data에서도 3d 객체를 탐지하기 위해 Frustum PointNets를 제안기존 연구들은 Images나 3D Voxels에서 natural 3D patterns를 학습→ Frustum PointNets는 raw point clouds에서 바로 학습하도록→3d bounding boxes를 정교하게 추정 가능→ occlusion, sparse해도 잘 예측할 수 O Introduction2D image understanding task(object detection, instance segmentation)는 많이 발전하지..

- Generative ModelPixelRNN, VAE, DCGAN, CycleGAN, ProGAN, StyleGAN, psp, ReStyle

- Image ClassificationLeNet, AlexNet, ZFNet, VGGNet, GoogleNet, ResNet, DenseNet, SENet, CBAM, NasNet, EfficientNet, ViT - Object detectionR-CNN, Fast R-CNN, Faster R-CNN, FPN, Mask R-CNN, YOLO, SSD, RetinaNet, EffiicientDet, FCOS, DETR - Domain AdaptationDANN, ADDA, CyCADA, Conditional UDA, FDA-MDT, MME

Pixel Recurrent Neural Networks Aaron van den Oord, Nal Kalchbrenner, Koray Kavukcuoglu Pixel Recurrent Neural Networks Modeling the distribution of natural images is a landmark problem in unsupervised learning. This task requires an image model that is at once expressive, tractable and scalable. We present a deep neural network that sequentially predicts the pixels in an i arxiv.org 0. Abstract..