The proceedings contain 101 papers. The topics discussed include: rotation invariant texture classification based on Gabor wavelets;group clustering and its visualization in group support systems;scene classification ...
详细信息
ISBN:
(纸本)9780819469526
The proceedings contain 101 papers. The topics discussed include: rotation invariant texture classification based on Gabor wavelets;group clustering and its visualization in group support systems;scene classification using low-level feature and intermediate feature;new method of 3D point reconstruction from monocular camera;probability output modeling for support vector machines;binocular stereo vision system design for lunar rover;shape classification using hidden Markov model and structural feature;research on target recognition techniques of radar networking based on fuzzy mathematics;novel cross correlation method for redshift determination of galaxy spectra;research on the city's water affairs dispatchment system based on rough sets theory;texture classification of aerial image based on Bayesian networks;and recognizing license plate character based on simplified PCNN.
The proceedings contain 63 papers. The topics discussed include: rural settlements extraction based on deep learning from high spatial resolution remote sensing imagery;G-CNN type recognition of typical aircraft based...
ISBN:
(纸本)9781510636378
The proceedings contain 63 papers. The topics discussed include: rural settlements extraction based on deep learning from high spatial resolution remote sensing imagery;G-CNN type recognition of typical aircraft based on target characteristics;semantic segmentation network combined with edge detection for building extraction in remote sensing images;multi-scales feature integration single shot multi-box detector on small object detection;a novel neural network approach for error transfer analysis in electronic instrument transformer;large-scale book page retrieval by deep hashing networks;and object direction estimation by constrained convolutional neural network.
The proceedings contain 100 papers. The topics discussed include: human recognition in a video network;the guaranteed cost switch control of BTT vehicle based on RBF-NN compensation;a novel method for evaluating the v...
ISBN:
(纸本)9780819478078
The proceedings contain 100 papers. The topics discussed include: human recognition in a video network;the guaranteed cost switch control of BTT vehicle based on RBF-NN compensation;a novel method for evaluating the validity of the visual attended regions based on SIFT descriptors;human body motion tracking based on quantum-inspired immune cloning algorithm;object detection with geometric context of keypoints described as lifetime;a combined feature latent semantic model for scene classification;spectral clustering with eigenvector selection based on ensemble ranking;to generate a finite element model of human thorax using the VCH dataset;three-dimensional building reconstruction using highly overlapped aerial images;study on human vision model of the multi-parameter correction factor;a hybrid registration approach in super-resolution reconstruction for visual surveillance application;and realistic generation of natural phenomena based on video synthesis.
The proceedings contain 57 papers. The topics discussed include: transfer estimation and the applications in data stream classification;a fast randomized clustering method based on a hypothetical potential field;a fin...
ISBN:
(纸本)9780819485786
The proceedings contain 57 papers. The topics discussed include: transfer estimation and the applications in data stream classification;a fast randomized clustering method based on a hypothetical potential field;a fingerprint matching algorithm based on probabilistic graphical model and three-tree model;personalized tag prediction via social influence in social networks;image super-resolution enhancement based on online learning and blind sparse decomposition;reference image generation based on three-dimensional realistic scene modeling for contour-based infrared ATR algorithm;extending the depth-of-field of incoherent imaging system through wave-front coding by using composite phase masks;a novel radiometric projector compensation algorithm based on Lambertian reflection model;improved Fourier descriptors in model-based pose estimation;and a method of frost observation based on intensity changing regularity simulation and texture analysis.
The proceedings contain 25 papers. The topics discussed include: accurate extraction of line-structured optical stripe centerlines under low exposure;research on grid inspection technology based on general knowledge e...
ISBN:
(纸本)9781510674950
The proceedings contain 25 papers. The topics discussed include: accurate extraction of line-structured optical stripe centerlines under low exposure;research on grid inspection technology based on general knowledge enhanced multi-modal large language models;a comparison between deep-learning models for scene recognition;binary network design for dedicated hardware circuits;evaluation of training datasets for typical target detection networks;optimal exposure selection of high dynamic range-based reflection compensation for printed circuit board reconstruction using structured light 3D measurement system;fall detection model based on Alphapose combined with LSTM and Lightgbm;and an image enhancement algorithm based on multi-scale Retinex theory to improve the images quality of sensors.
The proceedings contain 47 papers. The topics discussed include: change detection based on auto-encoder model for VHR images;a robust feature-based registration method of multimodal image using phase congruency and co...
ISBN:
(纸本)9780819498045
The proceedings contain 47 papers. The topics discussed include: change detection based on auto-encoder model for VHR images;a robust feature-based registration method of multimodal image using phase congruency and coherent point drift;the study on the feedback of large scale content-based video retrieval;BP network for atorvastatin effect evaluation from ultrasound images features classification;saliency detection based on manifold learning;the clustering algorithm of marine environment vector field data based on 'velocity-direction' histogram;salient region detection using Wasserstein distance measure based on nonlinear scale space;star sensor calibration based on interior-exterior parameters and error compensation;template match using local feature with view invariance;face recognition under variable illumination via sparse representation of patches;and statistics of the number of students in the classroom based on frequency domain analysis and seat detection.
Recent advances in hand-crafted neural architectures for visual recognition underscore the pressing need to explore architecture designs comprising diverse building blocks. Concurrently, neural architecture search (NA...
详细信息
Recent advances in hand-crafted neural architectures for visual recognition underscore the pressing need to explore architecture designs comprising diverse building blocks. Concurrently, neural architecture search (NAS) methods have gained traction as a means to alleviate human efforts. Nevertheless, the question of whether NAS methods can efficiently and effectively manage diversified search spaces featuring disparate candidates, such as Convolutional Neural Networks (CNNs) and transformers, remains an open question. In this work, we introduce a novel unsupervised NAS approach called BossNAS (Block-wisely Self-supervised Neural Architecture Search), which aims to address the problem of inaccurate predictive architecture ranking caused by a large weight-sharing space while mitigating potential ranking issue caused by biased supervision. To achieve this, we factorize the search space into blocks and introduce a novel self-supervised training scheme called Ensemble Bootstrapping, to train each block separately in an unsupervised manner. In the search phase, we propose an unsupervised Population-Centric Search, optimizing the candidate architecture towards the population center. Additionally, we enhance our NAS method by integrating masked image modeling and present BossNAS++ to overcome the lack of dense supervision in our block-wise self-supervised NAS. In BossNAS++, we introduce the training technique named Masked Ensemble Bootstrapping for block-wise supernet, accompanied by a Masked Population-Centric Search scheme to promote fairer architecture selection. Our family of models, discovered through BossNAS and BossNAS++, delivers impressive results across various search spaces and datasets. Our transformer model discovered by BossNAS++ attains a remarkable accuracy of 83.2% on ImageNet with only 10.5B MAdds, surpassing DeiT-B by 1.4% while maintaining a lower computation cost. Moreover, our approach excels in architecture rating accuracy, achieving Spearman correlat
The proceedings contain 525 papers. The topics discussed include: learning visual similarity measures for comparing never seen objects;a contextual dissimilarity measure for accurate and efficient image search;learnin...
详细信息
ISBN:
(纸本)1424411807
The proceedings contain 525 papers. The topics discussed include: learning visual similarity measures for comparing never seen objects;a contextual dissimilarity measure for accurate and efficient image search;learning local image descriptors;principal curvature-based region detector for object recognition;a benchmark for the comparison of 3-D motion segmentation algorithms;a nonparametric treatment for location/segmentation based visual tracking;learning gaussian conditional random fields for low-level vision;hierarchical structuring of data on manifolds;learning GMRF structures for spatial priors;element rearrangement for tensor-based subspace learning;unsupervised clustering using multi-resolution perceptual grouping;object tracking by asymmetric kernel mean shift with automatic scale and orientation selection;and free-form nonrigid image registration using generalized elastic nets.
The proceedings contain 95 papers. The topics discussed include: a novel level-independent fingerprint segmentation algorithm based on Fourier energy;towards an automated end-to-end face recognition system;a computati...
ISBN:
(纸本)9781601320438
The proceedings contain 95 papers. The topics discussed include: a novel level-independent fingerprint segmentation algorithm based on Fourier energy;towards an automated end-to-end face recognition system;a computationally efficient fingerprint segmentation algorithm using digital closed curves;grouping, segmentation and recognition of handwritten social security images;a multi-level component-based approach for face recognition;handwritten and typewritten word and character separation in unconstrained document images;recognition of handwritten Arabic writing with Kohonen networks;tracking multiple sports players for mobile display;genetic eye tracking for blinking;and target tracking and posture estimation of 3D objects by using particle filter and parametric eigenspace method.
暂无评论