RSS Feed - Blog
About: In DMNS source, five databases are used in slover.cpp and data_veh_layer.cpp, these images and databases are included in this file, except munich database. Changes:Initial Announcement on mloss.org.
|
About: Deep measuring net sequence(DMNS) is a sequence of three deep measuring nets, the later are deep fcn-based networks, directely output object category score, object orientation, location and scale simultaneously without any anchor boxes. DMNS acheived high accuracy in maneuvering target detection and geometrical measurements. Its average orientation error is less than 3.5 degree, loaction error less than 1.3 pixel, scale measuring error less than 10%, achieve a detection F1-score 96.5% in OAD, 91.8% in SVDS ,90.8% in Munich , 87.3% in OIRDS, outperforms SSD, Fater R-CNN, etc. Changes:Initial Announcement on mloss.org.
|
About: This database include 164 satellite iamges of different airports from google-earth, the first 110 images are used as training images, include 2337 aircrafts, the remained 54 images are used as test images, include 2206 aircrafts, each aircraft are labeled by two points and one number, indicating the positions of head and tail,and which point is the head. The labeled informations are recorded in two files: train.txt and test.txt, matlab is recommanded to be used for reading these data by import data tool. Changes:Initial Announcement on mloss.org.
|
About: The scancity of open database of small targets in remotesensing field has hindered the research in a lot. This open database includes 111 satellite images from google earth, 80 images are used as training set, contains 7862 vehicles. The other 31 images sre used as test set, contains 1635 vehicles. The positions of each vehicle are labeled by two points, which are recorded in "train.txt" and "test.txt" respectively. The label information is easily to be read by matlab tool. So such an open database is very useful for researchers in object detection fields. Changes:Initial Announcement on mloss.org.
|
About: The proposed hashing algorithm leverages the bootstrap sampling idea and integrates it with PCA, resulting in a new projection method called Bagging PCA Hashing. Changes:Initial Announcement on mloss.org.
|
About: This is an online hashing algorithm which can handle the stream data with low computational cost. Changes:Initial Announcement on mloss.org.
|
|
About: a tool for marking samples in images for database building, also including algorithm of LBP,HOG,and classifiers of SVM (six kernels), adaboost,BP and convolutional networks, extreme learning machine. Changes:Initial Announcement on mloss.org.
|
About: This MATLAB package provides the MLAPG algorithm proposed in our ICCV 2015 paper. It is efficient for PSD constrained metric learning, and also effective for person re-identification. For more details, please visit http://www.cbsr.ia.ac.cn/users/scliao/projects/mlapg/. Changes:Initial Announcement on mloss.org.
|
About: This algorithm is described in Deep Semantic Ranking Based Hashing for Multi-Label Image Retrieval. See https://github.com/zhaofang0627/cuda-convnet-for-hashing Changes:Initial Announcement on mloss.org.
|
About: This MATLAB package provides the Deep Quadratic Tree (DQT) and the Normalized Pixel Difference (NPD) based face detector training method proposed in our PAMI 2015 paper. It is fast, and effective for unconstrained face detection. For more details, please visit http://www.cbsr.ia.ac.cn/users/scliao/projects/npdface/. Changes:Initial Announcement on mloss.org.
|
About: Jie Gui et al., "How to estimate the regularization parameter for spectral regression discriminant analysis and its kernel version?", IEEE Transactions on Circuits and Systems for Video Technology, vol. 24, no. 2, pp. 211-223, 2014 Changes:Initial Announcement on mloss.org. |
About: Jie Gui, Zhenan Sun, Guangqi Hou, Tieniu Tan, "An optimal set of code words and correntropy for rotated least squares regression", International Joint Conference on Biometrics, 2014, pp. 1-6 Changes:Initial Announcement on mloss.org.
|
About: This MATLAB package provides the LOMO feature extraction and the XQDA metric learning algorithms proposed in our CVPR 2015 paper. It is fast, and effective for person re-identification. For more details, please visit http://www.cbsr.ia.ac.cn/users/scliao/projects/lomo_xqda/. Changes:Initial Announcement on mloss.org.
|
About: Jmlp is a java platform for both of the machine learning experiments and application. I have tested it on the window platform. But it should be applicable in the linux platform due to the cross-platform of Java language. It contains the classical classification algorithm (Discrete AdaBoost.MH, Real AdaBoost.MH, SVM, KNN, MCE,MLP,NB) and feature reduction(KPCA,PCA,Whiten) etc. Changes:Initial Announcement on mloss.org.
|
About: The auto-encoder based data clustering toolkit provides a quick start of clustering based on deep auto-encoder nets. This toolkit can cluster data in feature space with a deep nonlinear nets. Changes:Initial Announcement on mloss.org.
|
About: This is an exact implementation of Histogram of Oriented Gradient as mentioned in the paper by Dalal. Changes:Initial Announcement on mloss.org.
|
About: This provide a semi-supervised learning method based co-training for RGB-D object recognition. Besides, we evaluate four state-of-the-art feature learing method under the semi-supervised learning framework. Changes:Initial Announcement on mloss.org.
|
About: This is an unoptimized implementation of the RFD binary descriptor, which is published in the following paper. B. Fan, et al. Receptive Fields Selection for Binary Feature Description. IEEE Transaction on Image Processing, 2014. doi: http://dx.doi.org/10.1109/TIP.2014.2317981 Changes:Initial Announcement on mloss.org.
|
About: Kaiye Wang, Ran He, Wei Wang, Liang Wang, Tiuniu Tan. Learning Coupled Feature Spaces for Cross-modal Matching. In ICCV, 2013. Changes:Initial Announcement on mloss.org.
|