A method for the reconstruction of 3D shape and texture from integral photography (IP) images is presented. Sharing the same principles with stereoscopic-based object reconstruction, it offers increased robustness to noise and occlusions due to the unique characteristics of IP images. A coarse-to-fine approach is used, employing what we believe to be a novel grid refinement step in order to increase the quality of the reconstructed objects. The proposed method's properties include configurable depth accuracy and direct and seamless triangulation. We evaluate our method using synthetic data from a computer-simulated IP setup as well as real data from a simple yet effective digital IP setup. Experiments show reconstructed objects of high-quality indicating that IP can be a competitive modality for 3D object reconstruction.
An Autostereoscopic 3 0 viewing .system that operates on the principles of Integral PhotographyIP provides a imique sen.se of depth. full parallax and multi-view functionality. The inherent redundancy of these images results into great amounts of data that should be efficiently coded for transmission or storage operations. In this communication a method for efficient coding of such images is presented, targeting to 3 0 imaging but video applications. The method is based on common techniques broadly used in image compression and properly adjusted in order to take advantage of the spatial redundancies of IP images. The generalig and flexibility of the proposed approach along with the stability f a r a wide range of bit rates constitutes the basic characteristics of the technique. The proposed technique can be easily realized in sofmare or hardn'are for computer based or standalone applications.
Background
Various techniques have been proposed in the literature for phase and tool recognition from laparoscopic videos. In comparison, research in multilabel annotation of still frames is limited.
Methods
We describe a framework for multilabel annotation of images extracted from laparoscopic cholecystectomy (LC) videos based on multi‐instance multiple‐label learning. The image is considered as a bag of features extracted from local regions after coarse segmentation. A method based on variational Bayesian gaussian mixture models (VBGMM) is proposed for bag representation. Three techniques based on different feature extraction and bag representation models are employed for comparison.
Results
Four anatomical structures (abdominal wall, gallbladder, fat, and liver bed) and a tool‐like object (specimen bag) were annotated in 482 images. Our method achieved the best performance on single label accuracy: 0.87 (highest) and 0.69 (lowest). Moreover, the performance was >20% higher in terms of four multilabel classification error metrics (one‐error, ranking‐loss, hamming‐loss, and coverage).
Conclusions
Our approach provides an accurate and efficient image representation for multilabel classification of still images captured in LC.
Integral imaging (InIm) is a highly promising technique for the delivery of three-dimensional (3D) image content. During capturing, different views of an object are recorded as an array of elemental images (EIs), which form the integral image. High-resolution InIm requires sensors with increased resolution and produces huge amounts of highly correlated data. In an efficient encoding scheme for InIm compression both inter-EI and intra-EI correlations have to be properly exploited. We present an EI traversal scheme that maximizes the performance of InIm encoders by properly rearranging EIs to increase the intra-EI correlation of jointly coded EIs. This technique can be used to augment performance of both InIm specific and properly adapted general use encoder setups, used in InIm compression. An objective quality metric is also introduced for evaluating the effects of different traversal schemes on the encoder performance.
In most integral image analysis and processing tasks, accurate knowledge of the internal image structure is required. In this paper we present a robust framework for the accurate rectification of perspectively distorted integral images based on multiple line segment detection. The use of multiple line segments increases the overall fault tolerance of our framework providing strong statistical support for the rectification process. The proposed framework is used for the automatic rectification, metric correction, and rotation of distorted integral images. The performance of our framework is assessed over a number of integral images with varying scene complexity and noise levels.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.