Meitu Imaging & Vision LabMeitu Imaging & Vision Lab (MTlab) is the core R&D department of Meitu Inc., responsible for innovations in computer vision, machine learning, augmented reality and cloud computing. All Meitu products, which are recognized by billions of users worldwide, are powered by the MTlab.
Yansong Tang, Dajun Ding,Yongming Rao,Yu Zheng, Danyang Zhang, Lili Zhao,Jiwen Lu,Jie Zhou
CVPR, (2019): 1207-1216
In this paper we have introduced COIN, a new largescale dataset for comprehensive instructional video analysis
Cited by20BibtexViews31Links
0
0
North American Chapter of the Association for Computational Linguistics, (2019): 397-406
We propose a hierarchical Gated Recurrent Unit framework to tackle the utterance-level emotion recognition in dialogue systems, where the individual utterance embeddings are learned by the lower-level GRU and the contexts of utterances are captured by the upper-level GRU
Cited by9BibtexViews46Links
0
0
Jinzhan Su, Zhe Wang, Chunyuan Liao,Haibin Ling
CVPR Workshops, pp.0-0, (2019)
We proposed a high accuracy high speed face alignment algorithm, which effectively combines global shape regression and a cascaded set of local shape refinements
Cited by3BibtexViews8Links
0
0
national conference on artificial intelligence, (2019)
7 Conclusion In this paper, we propose a novel tensor decomposition based approach Centroid-based Multilayer Network Clustering to solve the multilayer networks clustering with multiple structures
Cited by2BibtexViews26Links
0
0
Gusi Te,Wei Hu, Amin Zheng,Zongming Guo
MM '18: ACM Multimedia Conference Seoul Republic of Korea October, 2018, pp.746-754, (2018)
We propose regularized graph convolutional neural network, a regularized graph convolutional neural network architecture that directly consumes irregular 3D point clouds
Cited by45BibtexViews2
0
0
ECCV, (2018)
We have proposed an interaction-aware self-attention which is inspired by PCA to further use non-local information in feature maps
Cited by32BibtexViews1Links
0
0
Suah Kim,Xiaochao Qu, Vasily Sachnev,Hyoung Joong Kim
IEEE Transactions on Circuits and Systems for Video Technology, no. 11 (2018): 1-1
Data hiding in an encrypted image is another use, where the original image recovery information is embedded as a reversible watermark
Cited by17BibtexViews1
0
0
IEEE Transactions on Circuits and Systems for Video Technology, no. 8 (2018): 2271-2284
Automatic image enhancement works in the following way: when a user uploads an image, the image is enhanced with no user interactions, but if the user is not satisfied with the enhancement at any time, the original image can be recovered
Cited by10BibtexViews2
0
0
Yue Gu,Xinyu Li, Kaixiang Huang, Shiyu Fu, Kangning Yang,Shuhong Chen, Moliang Zhou,Ivan Marsic
MM '18: ACM Multimedia Conference Seoul Republic of Korea October, 2018, (2018): 537-545
A word-level feature extraction strategy that can be widely used for emotion recognition, sentiment analysis, and associated applications
Cited by9BibtexViews1Links
0
0
IEEE Transactions on Image Processing, no. 7 (2018): 3332-3344
For the computer-generated depth images, we first injected noise to the depth images assuming that the pixels along the contours were corrupted by iid noise: for each edge of the contour, the corruption probability was fixed at δ
Cited by5BibtexViews7Links
0
0
AAAI, (2018)
We have proposed a new feature extraction method based on hierarchical Nonlinear Orthogonal Adaptive-Subspace Self-Organizing Map for action recognition
Cited by4BibtexViews7Links
0
0
Jie Huang,Pengfei Zhu, Mingrui Geng, Jiewen Ran, Xingguang Zhou, Chen Xing,Pengfei Wan,Xiangyang Ji
ECCV Workshops, pp.230-242, (2018)
We proposed the Range Scaling Global U-Net, a novel convolutional neural networks-based approach for perceptual image enhancement
Cited by2BibtexViews12Links
0
0
Chengyao Qian, Ting Liu, Hao Jiang, Zhe Wang,Pengfei Wang, Mingxin Guan, Biao Sun
arXiv: Computer Vision and Pattern Recognition, (2018)
In order to improve the efficiency, effectiveness, and accuracy of melanoma diagnosis, International Skin Imaging Collaboration In this report, we propose a method for lesion segmenprovides over 2,000 dermoscopic images of various skin tation
Cited by0BibtexViews1Links
0
0
Journal of Biomedical Informatics, (2018)
Using two deep textual views, a deep co-training framework is proposed based on factorization machines and ensemble learning
Cited by0BibtexViews13Links
0
0
ICCV, (2017): 5487-5495
We propose a new general tracking framework, named parallel tracking and verifying, which decomposes object tracking into two sub-tasks, tracking and verifying
Cited by176BibtexViews0Links
0
0
ICCV, pp.5468-5477, (2017)
In this paper we propose to use Structured Saliency Pattern for describing local saliency distributions
Cited by9BibtexViews8Links
0
0
ICRA, pp.4429-4436, (2017)
We have proposed an illumination-insensitive efficient second-order minimization for planar object tracking by introducing gradient orientations into direct visual tracking problem
Cited by9BibtexViews7Links
0
0
IEEE Transactions on Information Forensics and Security, no. 12 (2017): 2971-2985
We investigate the effectiveness of our proposed SIFT keypoint removal method Removal via Directed Graph Construction
Cited by7BibtexViews2Links
0
0
IEEE Trans. Image Processing, no. 6 (2016): 2896-2909
We proposed a computation-efficient algorithm for the image bit-depth enhancement problem, where an image of low bit-depth is reconstructed to one of high bit-depth
Cited by27BibtexViews9Links
0
0
IEEE Transactions on Information Forensics and Security, no. 8 (2016): 1722-1735
We have investigated the security of Scale Invariant Feature Transform against malicious attacks
Cited by15BibtexViews4Links
0
0
Keywords
DistortionImage CodingFeature ExtractionHistogramsImage Color AnalysisImage Edge DetectionOptimizationPayloadsSecurityAlgorithm Design And Analysis
Authors
Haibin Ling
Paper 9
Tao Wang
Paper 3
Congyan Lang
Paper 2
Pengfei Wan
Paper 2
Chunfeng Yuan
Paper 2
Haiqin Yang
Paper 2
Weiming Hu
Paper 2
Lili Zhao
Paper 2
Bing Li
Paper 2
Jiantao Zhou
Paper 2