If the input frame is a new video shot frame, image type of this frame is judged and the global depth gradient is assigned accordingly. Whether the input frame of video stream belongs to a new video shot is first detected. As the camera moves slightly within a video shot, the global depth gradient is changed slightly and it can be shared by frames in the same video shot to reduce the computational complexity and promote the temporal smoothness of the depth maps with little extra inaccuracy, which is proved by our experimental results in Section 3. However, these methods are not adaptable to different image types and may produce temporal flicker as they ignore the temporal coherence of depth maps between frames in the original 2D video.Īn input 2D video stream consists of a number of video shots and each video shot includes a sequence of frames taken using a single camera. use three depth scene models and color theory to produce a depth map. generated a depth map from motion information provided in H.264 bit streams.
assigned depth map by relative height based on a line tracing method and depth refinement filter. A depth map can be obtained from single monocular view through various depth cues including texture gradient, geometric perspective, focus/defocus, interposition, motion parallax, relative height or size and so on. Depth information can be depicted by a depth map, which is a gray-scale image whose value for the farthest pixel is 0 and the nearest pixel is 255. 2D video has a lack of depth information compared with 3D video. In the process of 2D-to-3D video conversion, depth map generation is an essential procedure and the latter procedure is relatively fixed. The experimental results prove that this novel method can adapt to different image types, reduce computational complexity and improve the temporal smoothness of generated 3D video. As input 2D video content consists of a number of video shots, the proposed algorithm reuses the global depth gradient of frames within the same video shot to generate time-coherent depth maps. Global depth gradient is computed according to image type while local depth refinement is related to color information. In our algorithm, a depth map is generated by combining global depth gradient and local depth refinement for each frame of 2D video input. Our study presents a novel 2D-to-3D video conversion algorithm which can be adopted in a 3D image sensor. Generally, 2D image sensor and 2D-to-3D conversion chip can compose a 3D image sensor.
3D image sensors are a new method to produce stereoscopic video content conveniently and at a low cost, and can thus meet the urgent demand for 3D videos in the 3D entertaiment market. 3D entertainment is getting more and more popular and can be found in many contexts, such as TV and home gaming equipment. In this paper, we propose a novel 2D-to-3D video conversion method for 3D entertainment applications.
0 kommentar(er)
