This paper provides a unified framework for the interrelated topics of action spotting, the spatiotemporal detection and localization of human actions in video, and action recognition, the classification of a given vi...
详细信息
This paper provides a unified framework for the interrelated topics of action spotting, the spatiotemporal detection and localization of human actions in video, and action recognition, the classification of a given video into one of several predefined categories. A novel compact local descriptor of video dynamics in the context of action spotting and recognition is introduced based on visual spacetime oriented energy measurements. This descriptor is efficiently computed directly from raw image intensity data and thereby forgoes the problems typically associated with flow-based features. Importantly: the descriptor allows for the comparison of the underlying dynamics of two spacetime video segments irrespective of spatial appearance, such as differences induced by clothing, and with robustness to clutter. An associated similarity measure is introduced that admits efficient exhaustive search for an action template, derived from a single exemplar video, across candidate video sequences. The general approach presented for action spotting and recognition is,amenable to efficient implementation, which is deemed critical for many important applications. For action spotting, details of a real-time GPU-based instantiation of the proposed approach are provided. Empirical evaluation of both action spotting and action recognition on challenging datasets suggests the efficacy of the proposed approach, with state-of-the-art performance documented on standard datasets.
A spectral algorithm for processing staggered-pulse repetition time (SPRT) signals in weather radar is introduced. It includes new approaches for ground clutter filter and hydrometeor spectral moments estimation. The ...
详细信息
A spectral algorithm for processing staggered-pulse repetition time (SPRT) signals in weather radar is introduced. It includes new approaches for ground clutter filter and hydrometeor spectral moments estimation. The algorithm uses ideas similar to GMAP but applied to non-uniform sampled signals. This work is focused on staggered sequences 2/3, but can be extended to other staggered sequences. Monte Carlo experiments were used to evaluate the performance of the spectral moments estimators for simulated weather signal, in scenarios with and without the presence of ground clutter. When clutter is present, a study using different clutter-to-signal ratios was carried out, showing that the method can deal with a wide range of situations and is appropriate for implementation in real scenarios. A comparison against GMAP-TD was performed, showing similar estimation results for both algorithms and a fivefold processing speed improvement for the proposed method. The performance was also validated using real weather data RMA-12 from a radar located in San Carlos de Bariloche, Argentina. The proposed algorithm has an easy implementation and is a good candidate for real-time implementations.
Parsimony, including sparsity and low rank, has been shown to successfully model data in numerous machine learning and signal processing tasks. Traditionally, such modeling approaches rely on an iterative algorithm th...
详细信息
Parsimony, including sparsity and low rank, has been shown to successfully model data in numerous machine learning and signal processing tasks. Traditionally, such modeling approaches rely on an iterative algorithm that minimizes an objective function with parsimony-promoting terms. The inherently sequential structure and data-dependent complexity and latency of iterative optimization constitute a major limitation in many applications requiring real-time performance or involving large-scale data. Another limitation encountered by these modeling techniques is the difficulty of their inclusion in discriminative learning scenarios. In this work, we propose to move the emphasis from the model to the pursuit algorithm, and develop a process-centric view of parsimonious modeling, in which a learned deterministic fixed-complexity pursuit process is used in lieu of iterative optimization. We show a principled way to construct learnable pursuit process architectures for structured sparse and robust low rank models, derived from the iteration of proximal descent algorithms. These architectures learn to approximate the exact parsimonious representation at a fraction of the complexity of the standard optimization methods. We also show that appropriate training regimes allow to naturally extend parsimonious models to discriminative settings. State-of-the-art results are demonstrated on several challenging problems in image and audio processing with several orders of magnitude speed-up compared to the exact optimization algorithms.
作者：
Alexandre, EPena, ASobreira, MUniv Alcala de Henares
Dept Signal Theory & Commun Dept Teor Senal & Comun Escuela Politecn Super Alcala De Henares 28805 Spain Univ Vigo
Dept Signal Theory & Commun Dept Teor Senal & Comun ETSE Telecommun Vigo 36310 Spain
This letter presents a method for estimating the quan-. tization noise introduced by a nonuniform quantizer, like those used in the family of MPEG-2/4 AAC audio coders. The method is generalized for the case of estima...
详细信息
This letter presents a method for estimating the quan-. tization noise introduced by a nonuniform quantizer, like those used in the family of MPEG-2/4 AAC audio coders. The method is generalized for the case of estimating the mean squared or the maximum value of the quantization noise. Its use will allow the bit-allocation algorithm to be adapted to different coding scenarios, depending on the available number of bits. Using the proposed method, it is possible to implement a loopless bit-allocation algorithm, without the need for using any kind of iteration loops. This helps to dramatically reduce the computational complexity of the bit-allocation algorithm, making it easier to implement in real-time applications where computational,power is limited.
We propose a local bar-shaped structure detector that works in realtime on high-resolution images. It is based on the Radon transform. Specifically in the muti-scale variant, which is especially fast because it works...
详细信息
ISBN:
(纸本)9781510650817;9781510650800
We propose a local bar-shaped structure detector that works in realtime on high-resolution images. It is based on the Radon transform. Specifically in the muti-scale variant, which is especially fast because it works in integer mathematics and does not use interpolation. The Radon transform conventionally works on the whole image, and not locally. In this paper we describe how by stopping at the early stages of the Radon transform we are able to locate structures locally. We also provide an evaluation of the performance of the algorithm running on the CPU, GPU and DSP of mobile devices to process at acquisition time the images coming from the device's camera.
暂无评论