The Experts below are selected from a list of 13560 Experts worldwide ranked by ideXlab platform

Heungyeung Shum - One of the best experts on this subject based on the ideXlab platform.

  • Blurred non Blurred Image alignment using sparseness prior
    International Conference on Computer Vision, 2007
    Co-Authors: Lu Yuan, Jian Sun, Long Quan, Heungyeung Shum
    Abstract:

    Aligning a pair of Blurred and non-Blurred Images is a prerequisite for many Image and video restoration and graphics applications. The traditional alignment methods such as direct and feature-based approaches cannot be used due to the presence of motion blur in one Image of the pair. In this paper, we present an effective and accurate alignment approach for a Blurred/non-Blurred Image pair. We exploit a statistical characteristic of the real blur kernel - the marginal distribution of kernel value is sparse. Using this sparseness prior, we can search the best alignment which produces the sparsest blur kernel. The search is carried out in scale space with a coarse-to-fine strategy for efficiency. Finally, we demonstrate the effectiveness of our algorithm for Image deblurring, video restoration, and Image matting.

  • Image deblurring with Blurred noisy Image pairs
    International Conference on Computer Graphics and Interactive Techniques, 2007
    Co-Authors: Lu Yuan, Jian Sun, Long Quan, Heungyeung Shum
    Abstract:

    Taking satisfactory photos under dim lighting conditions using a hand-held camera is challenging. If the camera is set to a long exposure time, the Image is Blurred due to camera shake. On the other hand, the Image is dark and noisy if it is taken with a short exposure time but with a high camera gain. By combining information extracted from both Blurred and noisy Images, however, we show in this paper how to produce a high quality Image that cannot be obtained by simply denoising the noisy Image, or deblurring the Blurred Image alone. Our approach is Image deblurring with the help of the noisy Image. First, both Images are used to estimate an accurate blur kernel, which otherwise is difficult to obtain from a single Blurred Image. Second, and again using both Images, a residual deconvolution is proposed to significantly reduce ringing artifacts inherent to Image deconvolution. Third, the remaining ringing artifacts in smooth Image regions are further suppressed by a gain-controlled deconvolution process. We demonstrate the effectiveness of our approach using a number of indoor and outdoor Images taken by off-the-shelf hand-held cameras in poor lighting environments.

  • Blurred no Blurred Image alignment using kernel sparseness prior
    International Conference on Computer Vision, 2007
    Co-Authors: Lu Yuan, Jian Sun, Long Quan, Heungyeung Shum, Harry Shum
    Abstract:

    Aligning a pair of Blurred and non-Blurred Images is a prerequisite for many Image and video restoration and graphics applications. The traditional alignment methods such as direct and feature-based approaches cannot be used due to the presence of motion blur in one Image of the pair. In this paper, we present an effective and accurate alignment approach for a Blurred/non-Blurred Image pair. We exploit a statistical characteristic of the real blur kernel – the marginal distribution of kernel value is sparse. Using this sparseness prior, we can search the best alignment which produces the sparsest blur kernel. The search is carried out in scale space with a coarse-to-fine strategy for efficiency. Finally, we demonstrate the effectiveness of our algorithm for Image deblurring, video restoration, and Image matting.

Lu Yuan - One of the best experts on this subject based on the ideXlab platform.

  • Blurred non Blurred Image alignment using sparseness prior
    International Conference on Computer Vision, 2007
    Co-Authors: Lu Yuan, Jian Sun, Long Quan, Heungyeung Shum
    Abstract:

    Aligning a pair of Blurred and non-Blurred Images is a prerequisite for many Image and video restoration and graphics applications. The traditional alignment methods such as direct and feature-based approaches cannot be used due to the presence of motion blur in one Image of the pair. In this paper, we present an effective and accurate alignment approach for a Blurred/non-Blurred Image pair. We exploit a statistical characteristic of the real blur kernel - the marginal distribution of kernel value is sparse. Using this sparseness prior, we can search the best alignment which produces the sparsest blur kernel. The search is carried out in scale space with a coarse-to-fine strategy for efficiency. Finally, we demonstrate the effectiveness of our algorithm for Image deblurring, video restoration, and Image matting.

  • Image deblurring with Blurred noisy Image pairs
    International Conference on Computer Graphics and Interactive Techniques, 2007
    Co-Authors: Lu Yuan, Jian Sun, Long Quan, Heungyeung Shum
    Abstract:

    Taking satisfactory photos under dim lighting conditions using a hand-held camera is challenging. If the camera is set to a long exposure time, the Image is Blurred due to camera shake. On the other hand, the Image is dark and noisy if it is taken with a short exposure time but with a high camera gain. By combining information extracted from both Blurred and noisy Images, however, we show in this paper how to produce a high quality Image that cannot be obtained by simply denoising the noisy Image, or deblurring the Blurred Image alone. Our approach is Image deblurring with the help of the noisy Image. First, both Images are used to estimate an accurate blur kernel, which otherwise is difficult to obtain from a single Blurred Image. Second, and again using both Images, a residual deconvolution is proposed to significantly reduce ringing artifacts inherent to Image deconvolution. Third, the remaining ringing artifacts in smooth Image regions are further suppressed by a gain-controlled deconvolution process. We demonstrate the effectiveness of our approach using a number of indoor and outdoor Images taken by off-the-shelf hand-held cameras in poor lighting environments.

  • Blurred no Blurred Image alignment using kernel sparseness prior
    International Conference on Computer Vision, 2007
    Co-Authors: Lu Yuan, Jian Sun, Long Quan, Heungyeung Shum, Harry Shum
    Abstract:

    Aligning a pair of Blurred and non-Blurred Images is a prerequisite for many Image and video restoration and graphics applications. The traditional alignment methods such as direct and feature-based approaches cannot be used due to the presence of motion blur in one Image of the pair. In this paper, we present an effective and accurate alignment approach for a Blurred/non-Blurred Image pair. We exploit a statistical characteristic of the real blur kernel – the marginal distribution of kernel value is sparse. Using this sparseness prior, we can search the best alignment which produces the sparsest blur kernel. The search is carried out in scale space with a coarse-to-fine strategy for efficiency. Finally, we demonstrate the effectiveness of our algorithm for Image deblurring, video restoration, and Image matting.

Dhruv Batra - One of the best experts on this subject based on the ideXlab platform.

  • human attention in visual question answering do humans and deep networks look at the same regions
    Computer Vision and Image Understanding, 2017
    Co-Authors: Abhishek Das, Harsh Agrawal, Lawrence C Zitnick, Devi Parikh, Dhruv Batra
    Abstract:

    Abstract We conduct large-scale studies on ‘human attention’ in Visual Question Answering (VQA) to understand where humans choose to look to answer questions about Images. We design and test multiple game-inspired novel attention-annotation interfaces that require the subject to sharpen regions of a Blurred Image to answer a question. Thus, we introduce the VQA-HAT (Human ATtention) dataset. We evaluate attention maps generated by state-of-the-art VQA models against human attention both qualitatively (via visualizations) and quantitatively (via rank-order correlation). Our experiments show that current attention models in VQA do not seem to be looking at the same regions as humans. Finally, we train VQA models with explicit attention supervision, and find that it improves VQA performance.

  • human attention in visual question answering do humans and deep networks look at the same regions
    arXiv: Machine Learning, 2016
    Co-Authors: Abhishek Das, Harsh Agrawal, Lawrence C Zitnick, Devi Parikh, Dhruv Batra
    Abstract:

    We conduct large-scale studies on `human attention' in Visual Question Answering (VQA) to understand where humans choose to look to answer questions about Images. We design and test multiple game-inspired novel attention-annotation interfaces that require the subject to sharpen regions of a Blurred Image to answer a question. Thus, we introduce the VQA-HAT (Human ATtention) dataset. We evaluate attention maps generated by state-of-the-art VQA models against human attention both qualitatively (via visualizations) and quantitatively (via rank-order correlation). Overall, our experiments show that current attention models in VQA do not seem to be looking at the same regions as humans.

Hueiyung Lin - One of the best experts on this subject based on the ideXlab platform.

  • vehicle speed detection from a single motion Blurred Image
    Image and Vision Computing, 2008
    Co-Authors: Hueiyung Lin, Chiahong Chang
    Abstract:

    An Image-based method for vehicle speed detection is presented. Conventional speed measurement techniques use radar- or laser-based devices, which are usually more expensive compared to a passive camera system. In this work, a single Image captured with vehicle motion is used for speed measurement. Due to the relative motion between the camera and a moving object during the camera exposure time, motion blur occurs in the dynamic region of the Image. It provides a visual cue for the speed measurement of a moving object. An approximate target region is first segmented and blur parameters are estimated from the motion Blurred subImage. The Image is then deBlurred and used to derive other parameters. Finally, the vehicle speed is calculated according to the imaging geometry, camera pose, and blur extent in the Image. Experiments have shown the estimated speeds within 5% of actual speeds for both local and highway traffic.

  • vehicle speed detection and identification from a single motion Blurred Image
    Workshop on Applications of Computer Vision, 2005
    Co-Authors: Hueiyung Lin
    Abstract:

    Motion blur is a result of finite acquisition time of practical cameras and the relative motion between the camera and moving objects. Traditionally, the Image degradations caused by motion blur are treated as undesirable artifacts and usually have to be removed before further processing. In this work, we propose a novel approach for vehicle speed detection based on a single motion Blurred Image as opposed to the most commonly used RADAR and LIDAR devices for traffic law enforcement. The motion blur parameters are estimated from a single motion Blurred Image and the length of motion blur is used for Image restoration. The restored Image is then used to obtain other parameters for vehicle speed estimation. The Images taken with the vehicle's license plates are used for both the assistance of Image restoration and the identification of the vehicle. We have established a link between the motion blur information of a 2D Image and the speed information of a moving object. Experiments have shown the results of less than 2% error for both local and highway traffic compared to video-based speed estimation methods

Shoujue Wang - One of the best experts on this subject based on the ideXlab platform.

  • a novel Image restoration algorithm based on high dimensional space geometry
    Lecture Notes in Computer Science, 2006
    Co-Authors: Shoujue Wang
    Abstract:

    A novel Image restoration approach based on high-dimensional space geometry is proposed, which is quite different from the existing traditional Image restoration techniques. It is based on the homeomorphisms and Principle of Homology Continuity (PHC), an Image is mapped to a point in high-dimensional space. Begin with the original Blurred Image, we get two further Blurred Images, then the restored Image can be obtained through the regressive curve derived from the three points which are mapped form the Images. Experiments have proved the availability of this Blurred-Blurred-restored algorithm, and the comparison with the classical Wiener Filter approach is presented in final.