Methods and Apparatuses of Depth Estimation from Focus Information

    公开(公告)号:US20220309696A1

    公开(公告)日:2022-09-29

    申请号:US17677365

    申请日:2022-02-22

    Applicant: MEDIATEK INC.

    Abstract: A depth estimation from focus method and system includes receiving input image data containing focus information, generating an intermediate attention map by an AI model, normalizing the intermediate attention map into a depth attention map via a normalization function, and deriving expected depth values for the input image data containing focus information from the depth attention map. The AI model for depth estimation can be trained unsupervisedly without ground truth depth maps. The AI model of some embodiments is a shared network estimating a depth map and reconstructing an AiF image from a set of images with different focus positions.

    Method and Apparatus of Video Compression for Pre-stitched Panoramic Contents

    公开(公告)号:US20180124312A1

    公开(公告)日:2018-05-03

    申请号:US15600954

    申请日:2017-05-22

    Applicant: MEDIATEK INC.

    Abstract: Methods and apparatus of compression for pre-stitched pictures captured by multiple cameras of a panoramic video capture device are disclosed. At the encoder side, stitching information associated with a stitching process to form the pre-stitched pictures is used to encode a current block according to embodiments of the present invention, where the stitching information comprises calibration data, matching results, seam position, blending level, sensor data, or a combination thereof. In one embodiment, the stitching information corresponds to matching results associated with a projection process, and projection-based Inter prediction is used to encode the current block by projecting a reference block in a reference pre-stitched picture to coordinates of the current block. In another embodiment, the stitching information corresponds to seam information associated with seam detection, and seam-based Inter prediction is used to encode the current block by utilizing the seam information.

    Method and Apparatus of Video Compression for Non-stitched Panoramic Contents

    公开(公告)号:US20170118475A1

    公开(公告)日:2017-04-27

    申请号:US15284390

    申请日:2016-10-03

    Applicant: MediaTek Inc.

    CPC classification number: H04N19/56

    Abstract: Methods and apparatus of compression for non-stitched pictures captured by multiple cameras of a panoramic video capture device are disclosed. According to one embodiment, the system uses a RIBC (Remapped Intra Block Copy) mode, where the block vector (BV) or BV predictor is remapped using calibration data to reduce the search range. The mapped BV or BVP is also more efficient for coding. A color scaling process can be used with the RIBC mode to compensate the color/brightness discrepancy between images from different cameras. A projection-based Inter prediction method is also disclosed. The projection-based Inter prediction method takes into account different perspectives between two images captured from different cameras. Transform matrix is applied to a block candidate to project the block candidate to a position of a target block. The projected block candidate is used as a predictor for the target block.

Patent Agency Ranking