Abstract: A method and a system for transmitting a live media stream to a plurality of nodes under a multicast tree in peer-to-peer (P2P) networks are provided. First, the reliability of each node is calculated according to a number of child nodes that can be supported by the node and an accumulated on-line time of the node. According to the number of child nodes and the reliability of each node, the nodes are rearranged in a database to establish an ideal reliable multicast tree. Finally, according to the reliability of each node, a location of each node in the multicast tree is practically adapted in sequence in a descending order, and a relative delay between nodes is weighted and compared, so a level difference between the adapted level of the node in the multicast tree and the level of the node in the ideal reliable multicast tree is the smallest.

Abstract: A central decoding controller and a central decoder controlling method are disclosed. A video stream is processed and transmitted via at least two parallel channels. The method comprises steps of: receiving a video key frame obtained by decoding an original video frame via a first channel, and a plurality of intra-description frames neighboring the video key frame; receiving a predictive video frame obtained by processing a prediction analysis and an error correction to the original video frame via a second channel parallel to the first channel; receiving a plurality of inter-description frames via the second channel, the inter-description frames neighboring the video key frame; calculating correlation of the video key frame, the intra-description frames, and the inter-description frames; and selecting the video key frame or the predictive video frame as an output frame according to the correlation result. The method can improve video quality under wireless transmission or unstable internet transmission.

Abstract: A human-environment interactive system and a portable device using the same are provided. The human-environment interactive system includes a ubiquitous computing system, at least one home-equipment, and the portable device. The portable device is suitable for two-way communicating with the portable device, and includes a RF transceiver, a browser, and a controller. When the portable device is taken into an environmental space, the controller actively receives a set of situational information from the ubiquitous computing system through the RF transceiver. The situational information includes a status information of at least one home-equipment, and a real time position information of the portable device. The controller displays this set of situational information in the browser for a user to perceive variations of the surrounding environment, so as to operate the at least one home-equipment through the browser.

Abstract: A method and a system for transmitting a live media stream to a plurality of nodes under a multicast tree in peer-to-peer (P2P) networks are provided. First, the reliability of each node is calculated according to a number of child nodes that can be supported by the node and an accumulated on-line time of the node. According to the number of child nodes and the reliability of each node, the nodes are rearranged in a database to establish an ideal reliable multicast tree. Finally, according to the reliability of each node, a location of each node in the multicast tree is practically adapted in sequence in a descending order, and a relative delay between nodes is weighted and compared, so a level difference between the adapted level of the node in the multicast tree and the level of the node in the ideal reliable multicast tree is the smallest.

Abstract: A method and apparatus for identifying visual content foregrounds, the method comprises steps of: determining a 3-D opening-by-reconstruction modest structure element (BO) and a 3-D closing-by-reconstruction modest structure element (BC); comparing an original image with an image obtained by performing MSOR operation to the original image with BO and an image obtained by performing MSCR operation to the original image with BC so as to generate an enhanced top-hat image and an enhanced bottom-hat image; and locating an overlap region between the enhanced top-hat image and the enhanced bottom-hat image. The overlap region forms a foreground identifying screen which is capable of identifying and extracting—refined for obtaining delicate foregrounds.

Abstract: A central decoding controller and a central decoder controlling method are disclosed. A video stream is processed and transmitted via at least two parallel channels. The method comprises steps of: receiving a video key frame obtained by decoding an original video frame via a first channel, and a plurality of intra-description frames neighboring the video key frame; receiving a predictive video frame obtained by processing a prediction analysis and an error correction to the original video frame via a second channel parallel to the first channel; receiving a plurality of inter-description frames via the second channel, the inter-description frames neighboring the video key frame; calculating correlation of the video key frame, the intra-description frames, and the inter-description frames; and selecting the video key frame or the predictive video frame as an output frame according to the correlation result. The method can improve video quality under wireless transmission or unstable internet transmission.

Abstract: A method and apparatus for identifying visual content foregrounds, the method comprises steps of: determining a 3-D opening-by-reconstruction modest structure element (BO) and a 3-D closing-by-reconstruction modest structure element (BC); comparing an original image with an image obtained by performing MSOR operation to the original image with BO and an image obtained by performing MSCR operation to the original image with BC so as to generate an enhanced top-hat image and an enhanced bottom-hat image; and locating an overlap region between the enhanced top-hat image and the enhanced bottom-hat image. The overlap region forms a foreground identifying screen which is capable of identifying and extracting—refined for obtaining delicate foregrounds.

Abstract: A background removal apparatus comprises a color normalization module that normalizes an original image, and an image segmentation module that segments the normalized image into several segmented regions. A background mesh generation module simulates the variation of pixel colors in background and generates an interpolated background mesh. A comparison module compares the normalized image and the interpolated background mesh to form a background mask by extracting the coherent regions between them. A refinement module uses a refined rule to determine a final background mask, and a background removal module through which a pure foreground image is obtained. The apparatus combines the efficiency of both color and spatial clustering, and improves the capabilities of current image segmentation method to perform background removal.

Abstract: An image retrieval system for retrieving image similarities from a database is proposed. The image retrieval system uses a universal query mechanism (UQM) to locate statistically silent common features among sample query images from different feature sets. The UQM also adjusts the weight factor for each feature to meet a user's query demand.

Abstract: A background removal apparatus comprises a color normalization module that normalizes an original image, and an image segmentation module that segments the normalized image into several segmented regions. A background mesh generation module simulates the variation of pixel colors in background and generates an interpolated background mesh. A comparison module compares the normalized image and the interpolated background mesh to form a background mask by extracting the coherent regions between them. A refinement module uses a refined rule to determine a final background mask, and a background removal module through which a pure foreground image is obtained. The apparatus combines the efficiency of both color and spatial clustering, and improves the capabilities of current image segmentation method to perform background removal.

Abstract: An image retrieval system for retrieving image similarities from a database is proposed. The image retrieval system uses a universal query mechanism (UQM) to locate statistically silent common features among sample query images from different feature sets. The UQM also adjusts the weight factor for each feature to meet a user's query demand.