Filters








14,450 Hits in 4.2 sec

Image annotation bykNN-sparse graph-based label propagation over noisily tagged web images

Jinhui Tang, Richang Hong, Shuicheng Yan, Tat-Seng Chua, Guo-Jun Qi, Ramesh Jain
2011 ACM Transactions on Intelligent Systems and Technology  
To annotate the images more accurately, we propose a novel k NN-sparse graph-based semi-supervised learning approach for harnessing the labeled and unlabeled data simultaneously.  ...  In this paper, we exploit the problem of annotating a large-scale image corpus by label propagation over noisily-tagged web images.  ...  In this paper, a novel k NN-sparse graph-based semi-supervised learning method with regularization on training labels is proposed to annotate images by label propagation over the noisily-tagged web images  ... 
doi:10.1145/1899412.1899418 fatcat:tcwxxjmycjguzhvtuqocwbfqje

Pedestrian Attribute Recognition At Far Distance

Yubin DENG, Ping Luo, Chen Change Loy, Xiaoou Tang
2014 Proceedings of the ACM International Conference on Multimedia - MM '14  
Second, we present the benchmark performance by SVM-based method and propose an alternative approach that exploits context of neighboring pedestrian images for improved attribute inference.  ...  We show that the large-scale dataset facilitates the learning of robust attribute detectors with good generalization performance.  ...  This is significant in a dataset with large appearance diversity and ambiguity and it demonstrates that graph regularization can improve attribute inference.  ... 
doi:10.1145/2647868.2654966 dblp:conf/mm/DENGLLT14 fatcat:ar6ynigjsfapdgqb6t6zy2aoye

Visual Semantic Information Pursuit: A Survey

Daqi Liu, Miroslaw Bober, Josef Kittler
2019 IEEE Transactions on Pattern Analysis and Machine Intelligence  
Surprisingly, a comprehensive review for this exciting area is still lacking.  ...  It is the core task of many different computer vision applications, such as object detection, visual semantic segmentation, visual relationship detection, or scene graph generation.  ...  It includes 10,000 images extracted from the Microsoft COCO dataset, in which 9,000 images are used for training and the previous unlabelled stuff pixels are further densely annotated with extra 91 class  ... 
doi:10.1109/tpami.2019.2950025 pmid:31675316 fatcat:ulzpshelorg5bm63wjgfd7pdxi

Active microscopic cellular image annotation by superposable graph transduction with imbalanced labels

Jun Wang, Shih-Fu Chang, Xiaobo Zhou, Stephen T. C. Wong
2008 2008 IEEE Conference on Computer Vision and Pattern Recognition  
We formulate this as a graph based transductive learning problem and develop a novel method for label propagation.  ...  However, manual annotation of cells and images in genome-wide studies is cost prohibitive.  ...  Chris Bakal for their valuable assistance and feedback. We also thank Mr. Wei Liu for the useful discussion on toy experiments and Mr. Zheng Yin for the help in collecting the cell images.  ... 
doi:10.1109/cvpr.2008.4587746 dblp:conf/cvpr/WangCZW08 fatcat:voyfhni65bhixcdm2wyr2bo6ne

Variational Bayesian Methods For Multimedia Problems

Zhaofu Chen, S. Derin Babacan, Rafael Molina, Aggelos K. Katsaggelos
2014 IEEE transactions on multimedia  
VB inference is flexible to be applied in different practical problems, yet is broad enough to subsume as its special cases several alternative inference approaches including Maximum A Posteriori (MAP)  ...  Specifically, both VB and EP are variational methods that minimize functionals based on the Kullback-Leibler (KL) divergence.  ...  APPENDIX C IMAGE CLASSIFICATION AND ANNOTATION WITH HIERARCHICAL BAYESIAN MODEL AND VARIATIONAL INFERENCE In this section we discuss an image classification and annotation problem, where the variational  ... 
doi:10.1109/tmm.2014.2307692 fatcat:4btc3ek37neulcmo72v6d3qlzm

Interactive Binary Image Segmentation with Edge Preservation [article]

Jianfeng Zhang, Liezhuo Zhang, Yuankai Teng, Xiaoping Zhang, Song Wang, Lili Ju
2018 arXiv   pre-print
Furthermore, we take a bilateral affinity approach for the pairwise term in order to preserve edge information and denoise.  ...  Binary image segmentation plays an important role in computer vision and has been widely used in many applications such as image and video editing, object extraction, and photo composition.  ...  For the comparison with FBS, we take the commonly used total variation (TV) module to show the influence of different regularizer on segmentation results.  ... 
arXiv:1809.03334v1 fatcat:r3s2tvsukvep7ffgemctwyadtm

Class-specific Gaussian-multinomial latent Dirichlet allocation for image annotation

Zhiming Qian, Ping Zhong, Runsheng Wang
2015 EURASIP Journal on Advances in Signal Processing  
An analysis of previous supervised LDA models shows that the topics discovered by generative LDA models are driven by general image regularities rather than the semantic regularities for image annotation  ...  is proposed for explicitly exploiting the manifold structures in the high-order image space.  ...  To annotate a new image, we first get the corresponding core tensor and then predict potential tags with Bayesian inference.  ... 
doi:10.1186/s13634-015-0224-z fatcat:jyy76ste6zc6xlgaunvysya2bq

Visual Semantic Information Pursuit: A Survey [article]

Daqi Liu, Miroslaw Bober, Josef Kittler
2019 arXiv   pre-print
However, a comprehensive review for this exciting area is still lacking.  ...  It is the core task of many different computer vision applications, such as object detection, visual semantic segmentation, visual relationship detection or scene graph generation.  ...  It includes 10,000 images extracted from the Microsoft COCO dataset, in which 9,000 images are used for training and the previous unlabelled stuff pixels are further densely annotated with extra 91 classes  ... 
arXiv:1903.05434v1 fatcat:lhpgeykuebb3new6ovu3lo424u

Joint Inference in Weakly-Annotated Image Datasets via Dense Correspondence [chapter]

Michael Rubinstein, Ce Liu, William T. Freeman
2016 Dense Image Correspondences for Computer Vision  
We present a principled framework for inferring pixel labels in weakly-annotated image datasets.  ...  This model requires significantly less labeled data and assists in resolving ambiguities by propagating inferred annotations from images with stronger local visual evidences to images with weaker local  ...  Third, unlike previous approaches which rely on large training sets of densely labeled images, our method can make do with significant less data by efficiently leveraging regularities and structures in  ... 
doi:10.1007/978-3-319-23048-1_11 fatcat:nxy3v53s3fhn5mn2crgnydlxe4

Joint Inference in Weakly-Annotated Image Datasets via Dense Correspondence

Michael Rubinstein, Ce Liu, William T. Freeman
2016 International Journal of Computer Vision  
We present a principled framework for inferring pixel labels in weakly-annotated image datasets.  ...  This model requires significantly less labeled data and assists in resolving ambiguities by propagating inferred annotations from images with stronger local visual evidences to images with weaker local  ...  Third, unlike previous approaches which rely on large training sets of densely labeled images, our method can make do with significant less data by efficiently leveraging regularities and structures in  ... 
doi:10.1007/s11263-016-0894-5 fatcat:tts2agsfyjekvbszjwifqnw544

Auto-clustering Output Layer: Automatic Learning of Latent Annotations in Neural Networks [article]

Ozsel Kilinc, Ismail Uysal
2017 arXiv   pre-print
Activity Regularization (GAR) technique.  ...  In this paper, we discuss a different type of semi-supervised setting: a coarse level of labeling is available for all observations but the model has to learn a fine level of latent annotation for each  ...  These ideas have also been extended for simultaneous image classification and annotation [18] , [19] .  ... 
arXiv:1702.08648v2 fatcat:x7sngafwbjgpbedqr2ffqzic6i

Semi-supervised topic modeling for image annotation

Yuanlong Shao, Yuan Zhou, Xiaofei He, Deng Cai, Hujun Bao
2009 Proceedings of the seventeen ACM international conference on Multimedia - MM '09  
We propose a novel technique for semi-supervised image annotation which introduces a harmonic regularizer based on the graph Laplacian of the data into the probabilistic semantic model for learning latent  ...  In this way, databases with very few labeled images can be annotated better than previous works.  ...  Figure 3 : 3 Annotation examples with top returned words. First line: ground truth annotations; Second line: our Laplacian regularized image annotation model; Third line: pLSA-Words model.  ... 
doi:10.1145/1631272.1631346 dblp:conf/mm/ShaoZHCB09 fatcat:32fznsbqcjdvxeu2qcf3rdnneu

An End-to-End Network for Generating Social Relationship Graphs [article]

Arushi Goel, Keng Teck Ma, Cheston Tan
2019 arXiv   pre-print
given input image.  ...  To achieve this, one computational approach for representing human relationships and attributes is to use an explicit knowledge graph, which allows for high-level reasoning.  ...  This gives us ground-truth annotations for single-body images. The context images are cropped from the full images using bounding box values of the people with relationship annotations.  ... 
arXiv:1903.09784v1 fatcat:misdg77vgzf5niwp6qpy5fxldy

Effective Refinement of Distinctive Analysis of the Facial Matrices for Automatic Face Annotation

Kavitha G L
2021 International Journal for Research in Applied Science and Engineering Technology  
We deal with real world images which contains numerous faces captioned with equivalent names, it may be wrongly annotated.  ...  This is a challenging task because of the very large appearance variation in the images, as well as the potential mismatch between images and their captions.  ...  This wrong annotation may happen due to the variation in the images and mismatch between the images and their captions.  ... 
doi:10.22214/ijraset.2021.34869 fatcat:a3jk7apkavgpjhht2nzbzex6qy

New Graph Structured Sparsity Model for Multi-label Image Annotations

Xiao Cai, Feiping Nie, Weidong Cai, Heng Huang
2013 2013 IEEE International Conference on Computer Vision  
To solve this problem, we derive an efficient optimization algorithm with proved convergence. We perform extensive experiments on six multi-label image annotation benchmark data sets.  ...  In multi-label image annotations, because each image is associated to multiple categories, the semantic terms (label classes) are not mutually exclusive.  ...  An example of label correlations for class membership inference. Both images have large regions with blue color, and it is hard to decide to annotate them with "sky" or "ocean".  ... 
doi:10.1109/iccv.2013.104 dblp:conf/iccv/CaiNCH13 fatcat:dvzrwohhhjajbo574tz6xfxlzm
« Previous Showing results 1 — 15 out of 14,450 results