Open unizard opened 7 years ago
Universal Correspondence Network, NIPS2016, C.B. Choy http://papers.nips.cc/paper/6486-universal-correspondence-network
Learning Discriminative and Transformation Covariant Local Feature Detectors, ICCV2017, Google, Columbia URL: http://openaccess.thecvf.com/content_cvpr_2017/papers/Zhang_Learning_Discriminative_and_CVPR_2017_paper.pdf git: https://github.com/ColumbiaDVMM/Spread-out_Local_Feature_Descriptor
Learning Spread-out Local Feature Descriptors, ICCV2017, Google, Columbia URL: http://openaccess.thecvf.com/content_ICCV_2017/papers/Zhang_Learning_Spread-Out_Local_ICCV_2017_paper.pdf Git: https://github.com/ColumbiaDVMM/Spread-out_Local_Feature_Descriptor Video
DeepCD: Learning Deep Complementary Descriptors for Patch Representations URL: http://openaccess.thecvf.com/content_ICCV_2017/papers/Yang_DeepCD_Learning_Deep_ICCV_2017_paper.pdf Git: https://github.com/shamangary/DeepCD
Learning Compact Geometric Features, ICCV2017, Berkeley, Intel http://openaccess.thecvf.com/content_ICCV_2017/papers/Khoury_Learning_Compact_Geometric_ICCV_2017_paper.pdf
Learning local feature descriptors with triplets and shallow convolutional neural networks, BMVC2016, Milokajczyk URL: http://www.bmva.org/bmvc/2016/papers/paper119/paper119.pdf Git: https://github.com/vbalnt/tfeat
Comparative Evaluation of Hand-Crafted and Learned Local Features, CVPR2017 ETH, Microsoft https://cvg.ethz.ch/research/local-feature-evaluation/schoenberger2017comparative.pdf git: https://github.com/ahojnnes/local-feature-evaluation
F-pointNet처럼.. 2D와 3D 특징량의 만남이 이루어져야 진정한 의미가 될듯. Semantic Learning과 같이 형태의 변형이 이루어져도 유사도가 높은 분위를 찾아내는 것처럼.
활용포인트를 잘 잡아야 한다. 요즘 같은 시절에 논문쓸때 응용못잡으면 망함
Large-Scale Image Retrieval with Attentive Deep Local Features, ICCV2017 노현우 https://arxiv.org/pdf/1612.06321.pdf
Unsupervised learning of object landmarks by factorized spatial embeddings, CVPR2017 oxford Vedaldi https://arxiv.org/pdf/1705.02193.pdf
SSD-6D: Making RGB-Based 3D Detection and 6D Pose Estimation Great Again, ICCV2017 http://openaccess.thecvf.com/content_ICCV_2017/papers/Kehl_SSD-6D_Making_RGB-Based_ICCV_2017_paper.pdf
NetVLAD: CNN architecture for weakly supervised place recognition, CVPR2016, Inria https://www.cv-foundation.org/openaccess/content_cvpr_2016/papers/Arandjelovic_NetVLAD_CNN_Architecture_CVPR_2016_paper.pdf git: http://www.di.ens.fr/willow/research/netvlad/
Learning local feature descriptors with triplets and shallow convolutional neural networks, BMVC2016 Comparative Evaluation of Hand-Crafted and Learned Local Features, CVPR2017 Universal Correspondence Network, NIPS2016 Learning Discriminative and Transformation Covariant Local Feature Detectors, ICCV2017 Learning Spread-out Local Feature Descriptors, ICCV2017 Learning Compact Geometric Features, ICCV2017