The Experts below are selected from a list of 42216 Experts worldwide ranked by ideXlab platform

Antoni B Chan - One of the best experts on this subject based on the ideXlab platform.

  • learning dynamic Memory networks for object tracking
    European Conference on Computer Vision, 2018
    Co-Authors: Tianyu Yang, Antoni B Chan
    Abstract:

    Template-matching methods for visual tracking have gained popularity recently due to their comparable performance and fast speed. However, they lack effective ways to adapt to changes in the target object’s appearance, making their tracking accuracy still far from state-of-the-art. In this paper, we propose a dynamic Memory network to adapt the template to the target’s appearance variations during tracking. An LSTM is used as a Memory controller, where the input is the search feature map and the outputs are the control signals for the reading and writing process of the Memory Block. As the location of the target is at first unknown in the search feature map, an attention mechanism is applied to concentrate the LSTM input on the potential target. To prevent aggressive model adaptivity, we apply gated residual template learning to control the amount of retrieved Memory that is used to combine with the initial template. Unlike tracking-by-detection methods where the object’s information is maintained by the weight parameters of neural networks, which requires expensive online fine-tuning to be adaptable, our tracker runs completely feed-forward and adapts to the target’s appearance changes by updating the external Memory. Moreover, unlike other tracking methods where the model capacity is fixed after offline training – the capacity of our tracker can be easily enlarged as the Memory requirements of a task increase, which is favorable for memorizing long-term object information. Extensive experiments on OTB and VOT demonstrates that our tracker MemTrack performs favorably against state-of-the-art tracking methods while retaining real-time speed of 50 fps.

  • learning dynamic Memory networks for object tracking
    European Conference on Computer Vision, 2018
    Co-Authors: Tianyu Yang, Antoni B Chan
    Abstract:

    Template-matching methods for visual tracking have gained popularity recently due to their comparable performance and fast speed. However, they lack effective ways to adapt to changes in the target object’s appearance, making their tracking accuracy still far from state-of-the-art. In this paper, we propose a dynamic Memory network to adapt the template to the target’s appearance variations during tracking. An LSTM is used as a Memory controller, where the input is the search feature map and the outputs are the control signals for the reading and writing process of the Memory Block. As the location of the target is at first unknown in the search feature map, an attention mechanism is applied to concentrate the LSTM input on the potential target. To prevent aggressive model adaptivity, we apply gated residual template learning to control the amount of retrieved Memory that is used to combine with the initial template. Unlike tracking-by-detection methods where the object’s information is maintained by the weight parameters of neural networks, which requires expensive online fine-tuning to be adaptable, our tracker runs completely feed-forward and adapts to the target’s appearance changes by updating the external Memory. Moreover, unlike other tracking methods where the model capacity is fixed after offline training – the capacity of our tracker can be easily enlarged as the Memory requirements of a task increase, which is favorable for memorizing long-term object information. Extensive experiments on OTB and VOT demonstrates that our tracker MemTrack performs favorably against state-of-the-art tracking methods while retaining real-time speed of 50 fps.

  • learning dynamic Memory networks for object tracking
    arXiv: Computer Vision and Pattern Recognition, 2018
    Co-Authors: Tianyu Yang, Antoni B Chan
    Abstract:

    Template-matching methods for visual tracking have gained popularity recently due to their comparable performance and fast speed. However, they lack effective ways to adapt to changes in the target object's appearance, making their tracking accuracy still far from state-of-the-art. In this paper, we propose a dynamic Memory network to adapt the template to the target's appearance variations during tracking. An LSTM is used as a Memory controller, where the input is the search feature map and the outputs are the control signals for the reading and writing process of the Memory Block. As the location of the target is at first unknown in the search feature map, an attention mechanism is applied to concentrate the LSTM input on the potential target. To prevent aggressive model adaptivity, we apply gated residual template learning to control the amount of retrieved Memory that is used to combine with the initial template. Unlike tracking-by-detection methods where the object's information is maintained by the weight parameters of neural networks, which requires expensive online fine-tuning to be adaptable, our tracker runs completely feed-forward and adapts to the target's appearance changes by updating the external Memory. Moreover, the capacity of our model is not determined by the network size as with other trackers -- the capacity can be easily enlarged as the Memory requirements of a task increase, which is favorable for memorizing long-term object information. Extensive experiments on OTB and VOT demonstrates that our tracker MemTrack performs favorably against state-of-the-art tracking methods while retaining real-time speed of 50 fps.

Tianyu Yang - One of the best experts on this subject based on the ideXlab platform.

  • learning dynamic Memory networks for object tracking
    European Conference on Computer Vision, 2018
    Co-Authors: Tianyu Yang, Antoni B Chan
    Abstract:

    Template-matching methods for visual tracking have gained popularity recently due to their comparable performance and fast speed. However, they lack effective ways to adapt to changes in the target object’s appearance, making their tracking accuracy still far from state-of-the-art. In this paper, we propose a dynamic Memory network to adapt the template to the target’s appearance variations during tracking. An LSTM is used as a Memory controller, where the input is the search feature map and the outputs are the control signals for the reading and writing process of the Memory Block. As the location of the target is at first unknown in the search feature map, an attention mechanism is applied to concentrate the LSTM input on the potential target. To prevent aggressive model adaptivity, we apply gated residual template learning to control the amount of retrieved Memory that is used to combine with the initial template. Unlike tracking-by-detection methods where the object’s information is maintained by the weight parameters of neural networks, which requires expensive online fine-tuning to be adaptable, our tracker runs completely feed-forward and adapts to the target’s appearance changes by updating the external Memory. Moreover, unlike other tracking methods where the model capacity is fixed after offline training – the capacity of our tracker can be easily enlarged as the Memory requirements of a task increase, which is favorable for memorizing long-term object information. Extensive experiments on OTB and VOT demonstrates that our tracker MemTrack performs favorably against state-of-the-art tracking methods while retaining real-time speed of 50 fps.

  • learning dynamic Memory networks for object tracking
    European Conference on Computer Vision, 2018
    Co-Authors: Tianyu Yang, Antoni B Chan
    Abstract:

    Template-matching methods for visual tracking have gained popularity recently due to their comparable performance and fast speed. However, they lack effective ways to adapt to changes in the target object’s appearance, making their tracking accuracy still far from state-of-the-art. In this paper, we propose a dynamic Memory network to adapt the template to the target’s appearance variations during tracking. An LSTM is used as a Memory controller, where the input is the search feature map and the outputs are the control signals for the reading and writing process of the Memory Block. As the location of the target is at first unknown in the search feature map, an attention mechanism is applied to concentrate the LSTM input on the potential target. To prevent aggressive model adaptivity, we apply gated residual template learning to control the amount of retrieved Memory that is used to combine with the initial template. Unlike tracking-by-detection methods where the object’s information is maintained by the weight parameters of neural networks, which requires expensive online fine-tuning to be adaptable, our tracker runs completely feed-forward and adapts to the target’s appearance changes by updating the external Memory. Moreover, unlike other tracking methods where the model capacity is fixed after offline training – the capacity of our tracker can be easily enlarged as the Memory requirements of a task increase, which is favorable for memorizing long-term object information. Extensive experiments on OTB and VOT demonstrates that our tracker MemTrack performs favorably against state-of-the-art tracking methods while retaining real-time speed of 50 fps.

  • learning dynamic Memory networks for object tracking
    arXiv: Computer Vision and Pattern Recognition, 2018
    Co-Authors: Tianyu Yang, Antoni B Chan
    Abstract:

    Template-matching methods for visual tracking have gained popularity recently due to their comparable performance and fast speed. However, they lack effective ways to adapt to changes in the target object's appearance, making their tracking accuracy still far from state-of-the-art. In this paper, we propose a dynamic Memory network to adapt the template to the target's appearance variations during tracking. An LSTM is used as a Memory controller, where the input is the search feature map and the outputs are the control signals for the reading and writing process of the Memory Block. As the location of the target is at first unknown in the search feature map, an attention mechanism is applied to concentrate the LSTM input on the potential target. To prevent aggressive model adaptivity, we apply gated residual template learning to control the amount of retrieved Memory that is used to combine with the initial template. Unlike tracking-by-detection methods where the object's information is maintained by the weight parameters of neural networks, which requires expensive online fine-tuning to be adaptable, our tracker runs completely feed-forward and adapts to the target's appearance changes by updating the external Memory. Moreover, the capacity of our model is not determined by the network size as with other trackers -- the capacity can be easily enlarged as the Memory requirements of a task increase, which is favorable for memorizing long-term object information. Extensive experiments on OTB and VOT demonstrates that our tracker MemTrack performs favorably against state-of-the-art tracking methods while retaining real-time speed of 50 fps.

Rakesh Kumar - One of the best experts on this subject based on the ideXlab platform.

  • Branch and data herding: Reducing control and Memory divergence for error-tolerant GPU applications
    2014
    Co-Authors: John Sartori, Rakesh Kumar
    Abstract:

    Abstract—Control and Memory divergence between threads within the same execution bundle, or warp, have been shown to cause significant performance bottlenecks for GPU applications. In this paper, we exploit the observation that many GPU applications exhibit error tolerance to propose branch and data herding. Branch herding eliminates control divergence by forcing all threads in a warp to take the same control path. Data herding eliminates Memory divergence by forcing each thread in a warp to load from the same Memory Block. To safely and efficiently support branch and data herding, we propose a static analysis and compiler framework to prevent exceptions when control and data errors are introduced, a profiling framework that aims to maximize performance while maintaining acceptable output quality, and hardware optimizations to improve the performance benefits of exploiting error tolerance through branch and data herding. Our software implementation of branch herding on NVIDIA GeForce GTX 480 improves performance by up to 34 % (13%, on average) for a suite of NVIDIA CUDA SDK and Parboil [16] benchmarks. Our hardware implementation of branch herding improves performance by up to 55 % (30%, on average). Data herding improves performance by up to 32 % (25%, on average). Observed output quality degradation is minimal for several applications that exhibit error tolerance, especially for visual computing applications. EDICS: Parallel Architectures and Design Techniques I

  • branch and data herding reducing control and Memory divergence for error tolerant gpu applications
    International Conference on Parallel Architectures and Compilation Techniques, 2012
    Co-Authors: John Sartori, Rakesh Kumar
    Abstract:

    Control and Memory divergence between threads in the same execution bundle, or warp, can significantly throttle the performance of GPU applications. We exploit the observation that many GPU applications exhibit error tolerance to propose branch and data herding. Branch herding eliminates control divergence by forcing all threads in a warp to take the same control path. Data herding eliminates Memory divergence by forcing each thread in a warp to load from the same Memory Block. To safely and efficiently support branch and data herding, we propose a static analysis and compiler framework to prevent exceptions when control and data errors are introduced, a profiling framework that aims to maximize performance while maintaining acceptable output quality, and hardware optimizations to improve the performance benefits of exploiting error tolerance through branch and data herding. Our software implementation of branch herding on NVIDIA GeForce GTX 480 improves performance by up to 34% (13%, on average) for a suite of NVIDIA CUDA SDK and Parboil [7] benchmarks. Our hardware implementation of branch herding improves performance by up to 55% (30%, on average). Data herding improves performance by up to 32% (25%, on average). Observed output quality degradation is minimal for several applications that exhibit error tolerance, especially for visual computing applications. For a more detailed exposition of this work, see [6].

Xiaoming Liu - One of the best experts on this subject based on the ideXlab platform.

  • memnet a persistent Memory network for image restoration
    International Conference on Computer Vision, 2017
    Co-Authors: Ying Tai, Jian Yang, Xiaoming Liu
    Abstract:

    Recently, very deep convolutional neural networks (CNNs) have been attracting considerable attention in image restoration. However, as the depth grows, the longterm dependency problem is rarely realized for these very deep models, which results in the prior states/layers having little influence on the subsequent ones. Motivated by the fact that human thoughts have persistency, we propose a very deep persistent Memory network (MemNet) that introduces a Memory Block, consisting of a recursive unit and a gate unit, to explicitly mine persistent Memory through an adaptive learning process. The recursive unit learns multi-level representations of the current state under different receptive fields. The representations and the outputs from the previous Memory Blocks are concatenated and sent to the gate unit, which adaptively controls how much of the previous states should be reserved, and decides how much of the current state should be stored. We apply MemNet to three image restoration tasks, i.e., image denosing, super-resolution and JPEG deBlocking. Comprehensive experiments demonstrate the necessity of the MemNet and its unanimous superiority on all three tasks over the state of the arts. Code is available at https://github.com/tyshiwo/MemNet.

  • memnet a persistent Memory network for image restoration
    arXiv: Computer Vision and Pattern Recognition, 2017
    Co-Authors: Ying Tai, Jian Yang, Xiaoming Liu
    Abstract:

    Recently, very deep convolutional neural networks (CNNs) have been attracting considerable attention in image restoration. However, as the depth grows, the long-term dependency problem is rarely realized for these very deep models, which results in the prior states/layers having little influence on the subsequent ones. Motivated by the fact that human thoughts have persistency, we propose a very deep persistent Memory network (MemNet) that introduces a Memory Block, consisting of a recursive unit and a gate unit, to explicitly mine persistent Memory through an adaptive learning process. The recursive unit learns multi-level representations of the current state under different receptive fields. The representations and the outputs from the previous Memory Blocks are concatenated and sent to the gate unit, which adaptively controls how much of the previous states should be reserved, and decides how much of the current state should be stored. We apply MemNet to three image restoration tasks, i.e., image denosing, super-resolution and JPEG deBlocking. Comprehensive experiments demonstrate the necessity of the MemNet and its unanimous superiority on all three tasks over the state of the arts. Code is available at this https URL.

Michael Lentmaier - One of the best experts on this subject based on the ideXlab platform.

  • the effect of coupling Memory and Block length on spatially coupled serially concatenated codes
    Vehicular Technology Conference, 2021
    Co-Authors: Mojtaba Mahdavi, Muhammad Umar Farooq, Liang Liu, Ove Edfors, Viktor Owall, Michael Lentmaier
    Abstract:

    Spatially coupled serially concatenated codes (SC-SCCs) are a class of spatially coupled turbo-like codes, which have a close-to-capacity performance and low error floor. In this paper, we perform a comprehensive design space exploration, revealing different aspects of SC-SCCs and discussing various design trade-offs. In particular, we investigate the impact of coupling Memory, Block length, decoding window size, and number of iterations on the performance, complexity, and latency of SC-SCCs. As a result, we propose design guidelines to make the code design independent of the Block length. By introducing a modified window decoding schedule, we are able to demonstrate that the Block length and coupling Memory can be exchanged flexibly without changing the latency and complexity of decoding and without performance loss. Thus, thanks to spatial coupling, a certain code strength and performance can be achieved by either a very small Block length or a large one, while the complexity and latency are fixed. Moreover, our results show that using higher coupling Memory with smaller Blocks can even improve the performance without increasing the latency and complexity. For all considered cases we observe that the performance of SC-SCCs is improved with respect to the uncoupled ensembles for a fixed latency and complexity.

  • the effect of coupling Memory and Block length on spatially coupled serially concatenated codes
    arXiv: Information Theory, 2020
    Co-Authors: Mojtaba Mahdavi, Muhammad Umar Farooq, Liang Liu, Ove Edfors, Viktor Owall, Michael Lentmaier
    Abstract:

    Spatially coupled serially concatenated codes (SC-SCCs) are a class of spatially coupled turbo-like codes, which have a close-to-capacity performance and low error floor. In this paper we investigate the impact of coupling Memory, Block length, decoding window size, and number of iterations on the performance, complexity, and latency of SC-SCCs. Several design tradeoffs are presented to see the relation between these parameters in a wide range. Also, our analysis provides design guidelines for SC-SCCs in different scenarios to make the code design independent of Block length. As a result, Block length and coupling Memory can be exchanged flexibly without changing the latency and complexity. Also, we observe that the performance of SC-SCCs is improved with respect to the uncoupled ensembles for a fixed latency and complexity.