In the recent past, there has been an increasing trend of using Blind Signal Separation (BSS) or Independent Component Analysis (ICA) algorithm for bio medical data, especially in prosthesis and Human computer Interac...
详细信息
On-line handwriting recognition has been a frontier area of research for the last few decades under the purview of patternrecognition. Word processing.turns to be a vexing experience even if it is with the assistance...
详细信息
Due to the high activation sparsity and use of accumulates (AC) instead of expensive multiply-and-accumulates (MAC), neuromorphic spiking neural networks (SNNs) have emerged as a promising low-power alternative to tra...
Next-generation reconnaissance systems (NGRS) will offer dynamic tasking of a menu of sensor modalities such as video, multi/hyper-spectral and polarization data. A key issue is how best to exploit these modes in time...
详细信息
The proceedings contain 96 papers. The topics discussed include: night-time traffic surveillance: a robust framework for multi-vehicle detection, classification and tracking;vehicle tracking using projective particle ...
ISBN:
(纸本)9780769537184
The proceedings contain 96 papers. The topics discussed include: night-time traffic surveillance: a robust framework for multi-vehicle detection, classification and tracking;vehicle tracking using projective particle filter;comparative evaluation of stationary foreground object detection algorithms based on background subtraction techniques;creating human activity recognition systems using Pareto-based multiobjective optimization;automatic gait recognition using weighted binary pattern on video;incremental EM for probabilistic latent semantic analysis on human action recognition;towards generic detection of unusual events in video surveillance;a people counting system based on face detection and tracking in a video;a pruning approach improving face identification systems;emerging trends in persistent surveillance information fusion;human body articulation for action recognition in video sequences;and multisensor fusion for monitoring elderly activities at home.
image analysis techniques to be used in experimental studies are presented in order to show that more objective and reproducible measures of subvisual information as related to textured cell images are possible. The f...
详细信息
ISBN:
(纸本)0780302168
image analysis techniques to be used in experimental studies are presented in order to show that more objective and reproducible measures of subvisual information as related to textured cell images are possible. The focus is on the development of methods to quantify the nuclear chromatin pattern and thus produce some cytodiagnostic clues. Two approaches for cell texture analysis have been developed, namely the Markovian and the runlength. Original experiments are designed for the application of these methods and their assessment both individually and in comparison to each other. They are assessed on the basis of their sensitivity to external noise and external variations, such as uneven illumination or/and inconsistent staining, their reproducibility after some normalization against such variations, and finally their computational complexity and power. Furthermore, different texture parameters from both approaches are combined and discriminant analysis is applied in order to select the best experimental feature subset resulting in an overall optimal cell classification rate.
Visual question answering is fundamentally compositional in nature-a question like where is the dog? shares substructure with questions like what color is the dog? and where is the cat? This paper seeks to simultaneou...
详细信息
ISBN:
(纸本)9781467388511
Visual question answering is fundamentally compositional in nature-a question like where is the dog? shares substructure with questions like what color is the dog? and where is the cat? This paper seeks to simultaneously exploit the representational capacity of deep networks and the compositional linguistic structure of questions. We describe a procedure for constructing and learning neural module networks, which compose collections of jointly-trained neural "modules" into deep networks for question answering. Our approach decomposes questions into their linguistic substructures, and uses these structures to dynamically instantiate modular networks (with reusable components for recognizing dogs, classifying colors, etc.). The resulting compound networks are jointly trained. We evaluate our approach on two challenging datasets for visual question answering, achieving state-of-the-art results on both the VQA natural image dataset and a new dataset of complex questions about abstract shapes.
One major challenge for 3D pose estimation from a single RGB image is the acquisition of sufficient training data. In particular, collecting large amounts of training data that contain unconstrained images and are ann...
详细信息
ISBN:
(纸本)9781467388511
One major challenge for 3D pose estimation from a single RGB image is the acquisition of sufficient training data. In particular, collecting large amounts of training data that contain unconstrained images and are annotated with accurate 3D poses is infeasible. We therefore propose to use two independent training sources. The first source consists of images with annotated 2D poses and the second source consists of accurate 3D motion capture data. To integrate both sources, we propose a dual-source approach that combines 2D pose estimation with efficient and robust 3D pose retrieval. In our experiments, we show that our approach achieves state-of-the-art results and is even competitive when the skeleton structure of the two sources differ substantially.
In the task of infrared weak and small target recognition, in order to improve the image quality and solve the problem of poor learning ability of convolutional neural network (CNN) due to the imbalance of positive an...
详细信息
Realistic rendering in real-time augmented reality applications leads one to consider physical interactions between real and virtual worlds. One of these interactions is mutual occlusions in the rendered viewpoint. Th...
详细信息
暂无评论