Abstract: An image quality improvement method and an image processing apparatus using the same are provided. Denoising filtering is performed to an original image by a filter to obtain a preliminary processing image. The preliminary processing image is input to a multi-stage convolutional network model to generate an optimization image through the multi-stage convolutional network model. The multi-stage convolutional network model includes multiple convolutional network sub-models, and these convolutional network sub-models respectively correspond to different network architectures.

Abstract: A saliency map generation method and an image processing system using the same are provided. The method includes the following steps. An original underwater image is received. A blurring processing is performed on the original underwater image to generate a defocus map. The defocus map is input to an auxiliary convolutional network model to obtain multiple feature maps of the defocus map. The original underwater image and the feature maps are input to a main convolutional network model to generate a saliency map of the original underwater image.

Abstract: A saliency map generation method and an image processing system using the same are provided. The method includes the following steps. An original underwater image is received. A blurring processing is performed on the original underwater image to generate a defocus map. The defocus map is input to an auxiliary convolutional network model to obtain multiple feature maps of the defocus map. The original underwater image and the feature maps are input to a main convolutional network model to generate a saliency map of the original underwater image.

Abstract: An image quality improvement method and an image processing apparatus using the same are provided. Denoising filtering is performed to an original image by a filter to obtain a preliminary processing image. The preliminary processing image is input to a multi-stage convolutional network model to generate an optimization image through the multi-stage convolutional network model. The multi-stage convolutional network model includes multiple convolutional network sub-models, and these convolutional network sub-models respectively correspond to different network architectures.

Abstract: The invention provides a method for depth estimation in image or video obtained from a distorting medium. In the method, a pixel blurriness map is calculated. A rough depth map is then determined from the pixel blurriness map while assuming depth in a small local patch is uniform. The rough depth map is then refined. The method can be implemented in imaging systems, such as cameras or imaging processing computers, and the distorting medium can be an underwater medium, haze, fog, low lighting, or a sandstorm, for example. Preferred embodiments determined the blurriness map by calculating a difference between an original image and multi-scale Gaussian-filtered images to estimate the pixel blurriness map.

Abstract: The invention provides a method for depth estimation in image or video obtained from a distorting medium. In the method, a pixel blurriness map is calculated. A rough depth map is then determined from the pixel blurriness map while assuming depth in a small local patch is uniform. The rough depth map is then refined. The method can be implemented in imaging systems, such as cameras or imaging processing computers, and the distorting medium can be an underwater medium, haze, fog, low lighting, or a sandstorm, for example. Preferred embodiments determined the blurriness map by calculating a difference between an original image and multi-scale Gaussian-filtered images to estimate the pixel blurriness map.

Abstract: A video editing method for a video editing device is provided. The method includes: obtaining an editing segment of a video, wherein the editing segment follows a first segment of the video and precedes a second segment of the video, the first segment includes a first picture, the second segment includes a second picture, and a short-term reference index of the second picture points to the first picture; setting a header of a third picture in the first segment, such that the first picture is moved from a short-term reference picture list to a long-term reference picture list when the header of the third picture is decoded; changing the short-term reference index of the second picture into a long-term reference index, wherein the long-term reference index points to the first picture in the long-term reference picture list; and encoding the video. Accordingly, the speed of editing the video is increased.

Abstract: A video editing method for a video editing device is provided. The method includes: obtaining an editing segment of a video, wherein the editing segment follows a first segment of the video and precedes a second segment of the video, the first segment includes a first picture, the second segment includes a second picture, and a short-term reference index of the second picture points to the first picture; setting a header of a third picture in the first segment, such that the first picture is moved from a short-term reference picture list to a long-term reference picture list when the header of the third picture is decoded; changing the short-term reference index of the second picture into a long-term reference index, wherein the long-term reference index points to the first picture in the long-term reference picture list; and encoding the video. Accordingly, the speed of editing the video is increased.

Abstract: A reference picture information communication interface implemented in a hardware/software integrated H.264 encoder/decoder is provided. The reference picture information communication interface includes a reference state register (RSR) and a reconstruct address register (RAR). The RSR has N bits, wherein N represents the maximum number of reference pictures supported by the H.264 encoder/decoder. The RSR is used by the hardware for notifying a software decoded picture buffer (DPB) controller about a current reference picture existence information. The RAR is used by the software for notifying the H.264 encoder/decoder about an address of a reconstruction information of a currently decoded picture in a memory space.

Abstract: A reference picture information communication interface implemented in a hardware/software integrated H.264 encoder/decoder is provided. The reference picture information communication interface includes a reference state register (RSR) and a reconstruct address register (RAR). The RSR has N bits, wherein N represents the maximum number of reference pictures supported by the H.264 encoder/decoder. The RSR is used by the hardware for notifying a software decoded picture buffer (DPB) controller about a current reference picture existence information. The RAR is used by the software for notifying the H.264 encoder/decoder about an address of a reconstruction information of a currently decoded picture in a memory space.