Abstract: A method for determining a foreground image and a background image, the method includes the following steps, generating a characteristic data of each of N sub-region images of in an interested region of N parent images, classifying the N sub-region images to image groups of in M image groups according to the characteristic data of each of the N sub-region images, obtaining a motion level of each of the M image groups according to a motion area of in the N sub-region images, determining whether each the image group belongs to a background image group or a foreground image group according to each the motion level of each the image group and an image quantity of in each the image group. The method can correctly determine a foreground image and a background image, even a foreground object stops moving and stays in a viewable range of an image apparatus.

Abstract: An image processing device comprises an image sensing module, comprising a lens, configured to receive light; an image sensing unit, configured to receive the light through the lens to generate a raw image; and an image processing unit; wherein the image processing unit processes the raw image to generate a first image; the image processing unit obtains an interested region according to a predetermined algorithm and generates a second image according to the interested region from the raw image or the first image.

Abstract: A method for determining a foreground image and a background image, the method includes the following steps, generating a characteristic data of each of N sub-region images of in an interested region of N parent images, classifying the N sub-region images to image groups of in M image groups according to the characteristic data of each of the N sub-region images, obtaining a motion level of each of the M image groups according to a motion area of in the N sub-region images, determining whether each the image group belongs to a background image group or a foreground image group according to each the motion level of each the image group and an image quantity of in each the image group. The method can correctly determine a foreground image and a background image, even a foreground object stops moving and stays in a viewable range of an image apparatus.

Abstract: A method and apparatus for occlusion area detection based on block difference associated with a motion vector and a predicted block difference are disclosed. For each current block of a frame, motion estimation is performed based on a temporally previous frame and a temporally subsequent frame. Based on the motion vector derived, two reference blocks of the current block are located in the temporally neighboring frames. The block difference between these two reference blocks is calculated for the current block. By comparing the block difference with a predicted block difference of the current block, the current block is determined to be an occlusion block or not accordingly. The predicted block difference is updated by averaging the block difference of neighboring blocks in a non-motion boundary area.

Abstract: A method and apparatus for foreground detection by designating one of two non-occlusion areas as the foreground area according to results of position matching using two given motion vectors. The detection is performed by matching the position of the current occlusion area in the current frame with one or more positions of a previous occlusion area in a previous frame. The matching operation is based on motion vectors determined for the neighboring areas adjacent to the current occlusion area. The determination of the motion vectors are based on two neighboring frames temporally located before and after the current frame. If the position of the current occlusion area is matched with the previous occlusion area by using one of the motion vectors, the neighboring area corresponding to the motion vector is designated as a foreground area.

Abstract: A method and apparatus for foreground detection by designating one of two non-occlusion areas as the foreground area according to results of position matching using two given motion vectors. The detection is performed by matching the position of the current occlusion area in the current frame with one or more positions of a previous occlusion area in a previous frame. The matching operation is based on motion vectors determined for the neighboring areas adjacent to the current occlusion area. The determination of the motion vectors are based on two neighboring frames temporally located before and after the current frame. If the position of the current occlusion area is matched with the previous occlusion area by using one of the motion vectors, the neighboring area corresponding to the motion vector is designated as a foreground area.

Abstract: A method and apparatus for occlusion area detection based on block difference associated with a motion vector and a predicted block difference are disclosed. For each current block of a frame, motion estimation is performed based on a temporally previous frame and a temporally subsequent frame. Based on the motion vector derived, two reference blocks of the current block are located in the temporally neighboring frames. The block difference between these two reference blocks is calculated for the current block. By comparing the block difference with a predicted block difference of the current block, the current block is determined to be an occlusion block or not accordingly. The predicted block difference is updated by averaging the block difference of neighboring blocks in a non-motion boundary area.

Abstract: A depth information generator includes a receiving circuit and a depth information generating block having a first depth information generating circuit included therein. The receiving circuit is arranged for receiving a multi-view video stream which transmits a plurality of images respectively corresponding to different views. The first depth information generating circuit is coupled to the receiving circuit, and arranged for generating a first depth information output by only processing part of the received images. In addition, a depth information generating method includes following steps: receiving a multi-view video stream which transmits a plurality of images respectively corresponding to different views; and generating a first depth information output by only processing part of the received images.

Abstract: A method and a device for de-interlacing a video signal having a field of interlaced scan lines are provided. The device comprises a direction engine and an interpolator. The direction engine is configured to calculate a de-interlacing cost with respect to a plurality of target pixels according to a first block and a second block of a plurality of pixels in the first and second interlaced scan lines respectively along each of a plurality of predetermined directions, and determine an interpolating direction for the target pixels among the predetermined directions according to the calculated de-interlacing cost of each of the predetermined directions. The interpolator interpolates the target pixels between the first and second interlaced scan lines along the interpolating direction.

Abstract: A method of optimizing a compression of a still image in JPEG 2000 format, which includes the steps of: (a) dividing an image into a predetermined numbers of tiles each having a predetermined numbers of tile pixels; (b) decomposing each of the tiles into a predetermined number of subbands by Discrete Wavelet Transform (DWT); (c) partitioning each of the subbands into a predetermined number of code-blocks which are arranged into a plurality of bit-planes respectively; (d) minimizing a total rate with a given predetermined total distortion of image so as to pre-determine an optimal truncation point for each of the code-blocks; (e) truncating a predetermined part of the DWT coefficients for each of the code-blocks so as to reduce computational resource; and (f) encoding each of the code-blocks by embedded block-coding to form an embedded code-stream having a plurality of code passes.

Abstract: The present invention provides a high-speed, memory efficient parallel coding technique for embedded block coding with optimized truncation (EBCOT) used in still image compression. Attributing to parallel processing method and structure, it processes a discrete wavelet transform (DWT) coefficient at a clock cycle without any state variable stored. Therefore, the need of state variable memory can be avoid and the external memory bandwidth can be reduced. With the same cost of chip-area and lower power consumption, the processing rate of this invention is several times higher than conventional schemes. Furthermore, the present invention processes 50 M coefficients per second at 100 MHz and can encode lossless HDTV 720 p resolution pictures at 30 fps in real time.

Abstract: The present invention provides a high-speed, memory efficient parallel coding technique for embedded block coding with optimized truncation (EBCOT) used in still image compression. Attributing to parallel processing method and structure, it processes a discrete wavelet transform (DWT) coefficient at a clock cycle without any state variable stored. Therefore, the need of state variable memory can be avoid and the external memory bandwidth can be reduced. With the same cost of chip-area and lower power consumption, the processing rate of this invention is several times higher than conventional schemes. Furthermore, the present invention processes 50 M coefficients per second at 100 MHz and can encode lossless HDTV 720p resolution pictures at 30 fps in real time.

Abstract: A method of optimizing a compression of a still image in JPEG 2000 format, which includes the steps of: (a) dividing an image into a predetermined numbers of tiles each having a predetermined numbers of tile pixels; (b) decomposing each of the tiles into a predetermined number of subbands by Discrete Wavelet Transform (DWT); (c) partitioning each of the subbands into a predetermined number of code-blocks which are arranged into a plurality of bit-planes respectively; (d) minimizing a total rate with a given predetermined total distortion of image so as to pre-determine an optimal truncation point for each of the code-blocks; (e) truncating a predetermined part of the DWT coefficients for each of the code-blocks so as to reduce computational resource; and (f) encoding each of the code-blocks by embedded block-coding to form an embedded code-stream having a plurality of code passes.

Abstract: The present invention provides a high-speed, memory efficient parallel coding technique for embedded block coding with optimized truncation (EBCOT) used in still image compression. Attributing to parallel processing method and structure, it processes a discrete wavelet transform (DWT) coefficient at a clock cycle without any state variable stored. Therefore, the need of state variable memory can be avoid and the external memory bandwidth can be reduced. With the same cost of chip-area and lower power consumption, the processing rate of this invention is several times higher than conventional schemes. Furthermore, the present invention processes 50 M coefficients per second at 100 MHz and can encode lossless HDTV 720p resolution pictures at 30 fps in real time.