IEEE Trans Pattern Anal Mach Intell - Learning AND-OR Templates for Object Recognition and Detection.

Tópicos

{ structur(1116) can(940) graph(676) }
{ imag(1947) propos(1133) code(1026) }
{ learn(2355) train(1041) set(1003) }
{ method(1219) similar(1157) match(930) }
{ motion(1329) object(1292) video(1091) }
{ method(1557) propos(1049) approach(1037) }
{ model(2220) cell(1177) simul(1124) }
{ cancer(2502) breast(956) screen(824) }
{ framework(1458) process(801) describ(734) }
{ visual(1396) interact(850) tool(830) }
{ take(945) account(800) differ(722) }
{ studi(1410) differ(1259) use(1210) }
{ model(2341) predict(2261) use(1141) }
{ gene(2352) biolog(1181) express(1162) }
{ use(2086) technolog(871) perceiv(783) }
{ model(3404) distribut(989) bayesian(671) }
{ imag(1057) registr(996) error(939) }
{ assess(1506) score(1403) qualiti(1306) }
{ surgeri(1148) surgic(1085) robot(1054) }
{ problem(2511) optim(1539) algorithm(950) }
{ extract(1171) text(1153) clinic(932) }
{ case(1353) use(1143) diagnosi(1136) }
{ perform(999) metric(946) measur(919) }
{ compound(1573) activ(1297) structur(1058) }
{ record(1888) medic(1808) patient(1693) }
{ health(3367) inform(1360) care(1135) }
{ research(1218) medic(880) student(794) }
{ model(2656) set(1616) predict(1553) }
{ cost(1906) reduc(1198) effect(832) }
{ first(2504) two(1366) second(1323) }
{ analysi(2126) use(1163) compon(1037) }
{ use(976) code(926) identifi(902) }
{ method(2212) result(1239) propos(1039) }
{ detect(2391) sensit(1101) algorithm(908) }
{ can(774) often(719) complex(702) }
{ data(1737) use(1416) pattern(1282) }
{ inform(2794) health(2639) internet(1427) }
{ system(1976) rule(880) can(841) }
{ measur(2081) correl(1212) valu(896) }
{ bind(1733) structur(1185) ligand(1036) }
{ sequenc(1873) structur(1644) protein(1328) }
{ featur(3375) classif(2383) classifi(1994) }
{ imag(2830) propos(1344) filter(1198) }
{ network(2748) neural(1063) input(814) }
{ imag(2675) segment(2577) method(1081) }
{ patient(2315) diseas(1263) diabet(1191) }
{ studi(2440) review(1878) systemat(933) }
{ treatment(1704) effect(941) patient(846) }
{ error(1145) method(1030) estim(1020) }
{ chang(1828) time(1643) increas(1301) }
{ concept(1167) ontolog(924) domain(897) }
{ clinic(1479) use(1117) guidelin(835) }
{ algorithm(1844) comput(1787) effici(935) }
{ data(1714) softwar(1251) tool(1186) }
{ design(1359) user(1324) use(1319) }
{ control(1307) perform(991) simul(935) }
{ care(1570) inform(1187) nurs(1089) }
{ general(901) number(790) one(736) }
{ method(984) reconstruct(947) comput(926) }
{ search(2224) databas(1162) retriev(909) }
{ featur(1941) imag(1645) propos(1176) }
{ howev(809) still(633) remain(590) }
{ data(3963) clinic(1234) research(1004) }
{ risk(3053) factor(974) diseas(938) }
{ research(1085) discuss(1038) issu(1018) }
{ system(1050) medic(1026) inform(1018) }
{ import(1318) role(1303) understand(862) }
{ perform(1367) use(1326) method(1137) }
{ studi(1119) effect(1106) posit(819) }
{ blood(1257) pressur(1144) flow(957) }
{ spatial(1525) area(1432) region(1030) }
{ model(3480) simul(1196) paramet(876) }
{ monitor(1329) mobil(1314) devic(1160) }
{ ehr(2073) health(1662) electron(1139) }
{ state(1844) use(1261) util(961) }
{ patient(2837) hospit(1953) medic(668) }
{ data(2317) use(1299) case(1017) }
{ age(1611) year(1155) adult(843) }
{ medic(1828) order(1363) alert(1069) }
{ signal(2180) analysi(812) frequenc(800) }
{ group(2977) signific(1463) compar(1072) }
{ sampl(1606) size(1419) use(1276) }
{ data(3008) multipl(1320) sourc(1022) }
{ intervent(3218) particip(2042) group(1664) }
{ activ(1138) subject(705) human(624) }
{ time(1939) patient(1703) rate(768) }
{ patient(1821) servic(1111) care(1106) }
{ can(981) present(881) function(850) }
{ health(1844) social(1437) communiti(874) }
{ high(1669) rate(1365) level(1280) }
{ use(1733) differ(960) four(931) }
{ drug(1928) target(777) effect(648) }
{ result(1111) use(1088) new(759) }
{ implement(1333) system(1263) develop(1122) }
{ survey(1388) particip(1329) question(1065) }
{ estim(2440) model(1874) function(577) }
{ decis(3086) make(1611) patient(1517) }
{ process(1125) use(805) approach(778) }
{ activ(1452) weight(1219) physic(1104) }
{ method(1969) cluster(1462) data(1082) }

Resumo

This paper presents a framework for unsupervised learning of a hierarchical reconfigurable image template - the AND-OR Template (AOT) for visual objects. The AOT includes: (1) hierarchical composition as ''AND'' nodes, (2) deformation and articulation of parts as geometric "OR" nodes, and (3) multiple ways of composition as structural "OR" nodes. The terminal nodes are hybrid image templates (HIT) \cite{hit} that are fully generative to the pixels. We show that both the structures and parameters of the AOT model can be learned in an unsupervised way from images using an information projection principle. The learning algorithm consists of two steps: i) a recursive block pursuit procedure to learn the hierarchical dictionary of primitives, parts and objects, and ii) a graph compression procedure to minimize model structure for better generalizability. We investigate the factors that influence how well the learning algorithm can identify the underlying AOT. And we propose a number of ways to evaluate the performance of the learned AOTs through both synthesized examples and real world images. Our model advances the state-of-the-art for object detection by improving the accuracy of template matching.

Resumo Limpo

paper present framework unsupervis learn hierarch reconfigur imag templat andor templat aot visual object aot includ hierarch composit node deform articul part geometr node multipl way composit structur node termin node hybrid imag templat hit citehit fulli generat pixel show structur paramet aot model can learn unsupervis way imag use inform project principl learn algorithm consist two step recurs block pursuit procedur learn hierarch dictionari primit part object ii graph compress procedur minim model structur better generaliz investig factor influenc well learn algorithm can identifi under aot propos number way evalu perform learn aot synthes exampl real world imag model advanc stateoftheart object detect improv accuraci templat match

Resumos Similares

IEEE Trans Image Process - Multiresolution graph Fourier transform for compression of piecewise smooth images. ( 0,791038391591439 )
IEEE Trans Image Process - Coding order decision of B frames for rate-distortion performance improvement in single-view video and multiview video coding. ( 0,788570348155252 )
IEEE Trans Image Process - A sparse embedding and least variance encoding approach to hashing. ( 0,728231293983051 )
IEEE Trans Image Process - Progressive significance map and its application to error-resilient image transmission. ( 0,726725680919933 )
IEEE Trans Image Process - Graph regularized sparse coding for image representation. ( 0,726019515758327 )
IEEE Trans Image Process - Learning-based hierarchical graph for unsupervised matting and foreground estimation. ( 0,712862899460656 )
IEEE Trans Image Process - Adaptive compressed sensing recovery utilizing the property of signal's autocorrelations. ( 0,711821680445085 )
IEEE Trans Image Process - Interactive streaming of stored multiview video using redundant frame structures. ( 0,711631335030457 )
IEEE Trans Vis Comput Graph - Link Conditions for Simplifying Meshes with Embedded Structures. ( 0,705779334500875 )
IEEE Trans Image Process - Connected filtering based on multivalued component-trees. ( 0,704591027933932 )
Neural Comput - Temporal order detection and coding in nervous systems. ( 0,687440211067488 )
Comput. Biol. Med. - Finding splitting lines for touching cell nuclei with a shortest path algorithm. ( 0,682697719226389 )
IEEE Trans Image Process - Histogram contextualization. ( 0,677734368575964 )
IEEE Trans Image Process - Image super-resolution via sparse representation. ( 0,674243576302321 )
IEEE Trans Image Process - Stereo matching and view interpolation based on image domain triangulation. ( 0,668671963099484 )
IEEE Trans Vis Comput Graph - Visual Analysis of Large Graphs Using (X,Y)-clustering and Hybrid Visualizations. ( 0,667328412731801 )
IEEE Trans Image Process - Color-guided depth recovery from RGB-D data using an adaptive autoregressive model. ( 0,665822968977919 )
IEEE Trans Pattern Anal Mach Intell - Graph Regularized Nonnegative Matrix Factorization for Data Representation. ( 0,663359001938646 )
IEEE Trans Image Process - Context coding of depth map images under the piecewise-constant image model representation. ( 0,661293586239761 )
IEEE Trans Image Process - Absolute depth estimation from a single defocused image. ( 0,654479439917738 )
IEEE Trans Image Process - Optimizing the hierarchical prediction and coding in HEVC for surveillance and conference videos with background modeling. ( 0,64959976233451 )
IEEE Trans Image Process - Learning component-level sparse representation for image and video categorization. ( 0,647132951824985 )
IEEE Trans Image Process - Perceptual video coding based on SSIM-inspired divisive normalization. ( 0,64398061560497 )
IEEE Trans Image Process - Image deblurring and super-resolution by adaptive sparse domain selection and adaptive regularization. ( 0,642320431650625 )
IEEE Trans Image Process - Hyperspectral image segmentation using a new spectral unmixing-based binary partition tree representation. ( 0,641551229199738 )
IEEE Trans Pattern Anal Mach Intell - Tensor Sparse Coding for Positive Definite Matrices. ( 0,641444177952858 )
IEEE Trans Image Process - Nonlocally centralized sparse representation for image restoration. ( 0,636980888954381 )
IEEE Trans Image Process - Real-time scalable depth sensing with hybrid structured light illumination. ( 0,633441142745814 )
IEEE Trans Image Process - Model-assisted adaptive recovery of compressed sensing with imaging applications. ( 0,631790900652293 )
IEEE Trans Vis Comput Graph - Bas-Relief Modeling from Normal Images with Intuitive Styles. ( 0,630464359110271 )
IEEE Trans Image Process - A Probabilistic Associative Model for Segmenting Weakly-Supervised Images. ( 0,629921087615698 )
IEEE Trans Image Process - Binary compressed imaging. ( 0,629083397421038 )
IEEE Trans Image Process - Sliding-window raptor codes for efficient scalable wireless video broadcasting with unequal loss protection. ( 0,627339753542513 )
J. Comput. Biol. - Pathset graphs: a novel approach for comprehensive utilization of paired reads in genome assembly. ( 0,624631590193452 )
IEEE Trans Image Process - Adaptive sensing and optimal power allocation for wireless video sensors with sigma-delta imager. ( 0,624143874039992 )
J. Comput. Biol. - The approximability of shortest path-based graph orientations of protein-protein interaction networks. ( 0,623783388282218 )
IEEE Trans Vis Comput Graph - Flow Visualization with Quantified Spatial and Temporal Errors Using Edge Maps. ( 0,623744239330017 )
IEEE Trans Image Process - Spatially adaptive block-based super-resolution. ( 0,61851935032508 )
IEEE Trans Vis Comput Graph - The Design Space of Implicit Hierarchy Visualization: A Survey. ( 0,616096407615469 )
IEEE Trans Image Process - Web and personal image annotation by mining label correlation with relaxed visual graph embedding. ( 0,613951455416338 )
IEEE Trans Image Process - Dual-geometric neighbor embedding for image super resolution with sparse tensor. ( 0,613006669394455 )
IEEE Trans Image Process - MIQM: a multicamera image quality measure. ( 0,612346277671148 )
IEEE Trans Image Process - Hyperspectral image representation and processing with binary partition trees. ( 0,611299986227939 )
IEEE Trans Pattern Anal Mach Intell - Building Development Monitoring in Multitemporal Remotely Sensed Image Pairs with Stochastic Birth-Death Dynamics. ( 0,609796726746456 )
IEEE Trans Pattern Anal Mach Intell - Angular Embedding: A Robust Quadratic Criterion. ( 0,60924391252209 )
IEEE Trans Image Process - Overall well focused catadioptric image acquisition with multi-focal images: a model based method. ( 0,608344462361738 )
IEEE Trans Image Process - Fast l1-minimization algorithms for robust face recognition. ( 0,60829108479075 )
IEEE Trans Image Process - Joint registration and super-resolution with omnidirectional images. ( 0,60634085372757 )
IEEE Trans Image Process - Image fusion using higher order singular value decomposition. ( 0,605624406928565 )
IEEE Trans Image Process - Principle of superposition by direction images. ( 0,60512824318369 )
IEEE Trans Image Process - Automatic view synthesis by image-domain-warping. ( 0,604413205618481 )
Comput. Biol. Med. - An efficient compression scheme for 4-D medical images using hierarchical vector quantization and motion compensation. ( 0,603053095237586 )
Neural Comput - Multilabel classification with principal label space transformation. ( 0,602221467428995 )
Neural Comput - Intrinsic graph structure estimation using graph Laplacian. ( 0,601975090937407 )
IEEE Trans Image Process - A novel fractal image compression scheme with block classification and sorting based on Pearson's correlation coefficient. ( 0,601612932255333 )
IEEE Trans Image Process - On dependent bit allocation for multiview image coding with depth-image-based rendering. ( 0,599763444610961 )
IEEE Trans Pattern Anal Mach Intell - Consistent Latent Position Estimation and Vertex Classification for Random Dot Product Graphs. ( 0,598371909527836 )
IEEE Trans Image Process - Video quality pooling adaptive to perceptual distortion severity. ( 0,596110455045936 )
IEEE Trans Image Process - SNMFCA: supervised NMF-based image classification and annotation. ( 0,595546081325318 )
IEEE Trans Image Process - Variational segmentation of vector-valued images with gradient vector flow. ( 0,594973410837018 )
IEEE Trans Image Process - A new hybrid method for image approximation using the easy path wavelet transform. ( 0,593284696820703 )
IEEE Trans Image Process - Learning doubly sparse transforms for images. ( 0,593187505360451 )
IEEE Trans Image Process - A psychovisual quality metric in free-energy principle. ( 0,59263368297633 )
IEEE Trans Image Process - Group-based sparse representation for image restoration. ( 0,592396575250758 )
IEEE Trans Image Process - Enhanced compressed sensing recovery with level set normals. ( 0,592040929671147 )
IEEE Trans Image Process - On the construction of topology-preserving deformation fields. ( 0,590805141626382 )
IEEE Trans Image Process - A novel joint data-hiding and compression scheme based on SMVQ and image inpainting. ( 0,588015014852399 )
IEEE Trans Image Process - Learning-based, automatic 2D-to-3D image and video conversion. ( 0,586114930551323 )
IEEE Trans Image Process - Distributed multiple description video coding on packet loss channels. ( 0,585240873623946 )
Comput Math Methods Med - Dimensionality reduction by supervised neighbor embedding using laplacian search. ( 0,584497001756726 )
Comput Biol Chem - On topological indices for small RNA graphs. ( 0,584347036508827 )
IEEE Trans Vis Comput Graph - Visualizing the Evolution and Interaction of Vortices and Shear Layers in Time-Dependent 3D Flow. ( 0,582827125227642 )
IEEE Trans Image Process - Correspondence map-aided neighbor embedding for image intra prediction. ( 0,582228600955517 )
IEEE Trans Image Process - Compressive sensing of sparse tensors. ( 0,581642517249537 )
J. Comput. Biol. - Maximum parsimony, substitution model, and probability phylogenetic trees. ( 0,581250216393416 )
IEEE Trans Image Process - Dictionary learning for stereo image representation. ( 0,578300011314188 )
IEEE Trans Image Process - Arbitrarily shaped motion prediction for depth video compression using arithmetic edge coding. ( 0,57650984186499 )
IEEE Trans Image Process - Directional filtering transform for image/intra-frame compression. ( 0,576182525906684 )
IEEE Trans Image Process - High efficiency 3D video coding using new tools based on view synthesis. ( 0,575508511765626 )
IEEE Trans Image Process - Fingerprint Compression Based on Sparse Representation. ( 0,574479887558291 )
IEEE Trans Pattern Anal Mach Intell - Multi-Commodity Network Flow for Tracking Multiple People. ( 0,574299006570191 )
Comput Methods Programs Biomed - TreeVis: a MATLAB-based tool for tree visualization. ( 0,573461205686079 )
IEEE Trans Vis Comput Graph - Animating Wrinkles by Example on Non-Skinned Cloth. ( 0,573036891010382 )
IEEE Trans Image Process - Constrained and dimensionality-independent path openings. ( 0,569890951987732 )
IEEE Trans Image Process - An innovative lossless compression method for discrete-color images. ( 0,56982891784459 )
IEEE Trans Vis Comput Graph - Guided Multiview Ray Tracing for Fast Auralization. ( 0,568174939814284 )
IEEE Trans Image Process - Scalable coding of encrypted images. ( 0,56791526866954 )
IEEE Trans Image Process - Advanced screen content coding using color table and index map. ( 0,56738088054921 )
IEEE Trans Pattern Anal Mach Intell - C^4: Exploring Multiple Solutions in Graphical Models by Cluster Sampling. ( 0,567370369452137 )
IEEE Trans Vis Comput Graph - Joint Contour Nets. ( 0,567170374665008 )
IEEE Trans Image Process - Watermarking of free-view video. ( 0,56675507951312 )
IEEE Trans Image Process - Plenoptic layer-based modeling for image based rendering. ( 0,565072272971476 )
IEEE Trans Image Process - Rate-distortion optimized rate control for depth map-based 3-D video coding. ( 0,564987591078656 )
IEEE Trans Image Process - Single-image super-resolution reconstruction via learned geometric dictionaries and clustered sparse coding. ( 0,564551476382602 )
IEEE Trans Image Process - Resolution scalable image coding with reversible cellular automata. ( 0,564157002480016 )
IEEE Trans Image Process - Multiview-video-plus-depth coding based on the advanced video coding standard. ( 0,563845801330789 )
IEEE Trans Image Process - Regularized robust coding for face recognition. ( 0,56330431421766 )
IEEE Trans Image Process - Jointly optimized spatial prediction and block transform for video and image coding. ( 0,561719467254824 )
IEEE Trans Image Process - 3-D curvilinear structure detection filter via structure-ball analysis. ( 0,560419462654879 )
IEEE Trans Image Process - Monocular depth ordering using T-junctions and convexity occlusion cues. ( 0,559116574860802 )