-
SIRfyN: Single Image Relighting from your Neighbors
Authors:
D. A. Forsyth,
Anand Bhattad,
Pranav Asthana,
Yuanyi Zhong,
Yuxiong Wang
Abstract:
We show how to relight a scene, depicted in a single image, such that (a) the overall shading has changed and (b) the resulting image looks like a natural image of that scene. Applications for such a procedure include generating training data and building authoring environments. Naive methods for doing this fail. One reason is that shading and albedo are quite strongly related; for example, sharp…
▽ More
We show how to relight a scene, depicted in a single image, such that (a) the overall shading has changed and (b) the resulting image looks like a natural image of that scene. Applications for such a procedure include generating training data and building authoring environments. Naive methods for doing this fail. One reason is that shading and albedo are quite strongly related; for example, sharp boundaries in shading tend to appear at depth discontinuities, which usually apparent in albedo. The same scene can be lit in different ways, and established theory shows the different lightings form a cone (the illumination cone). Novel theory shows that one can use similar scenes to estimate the different lightings that apply to a given scene, with bounded expected error. Our method exploits this theory to estimate a representation of the available lighting fields in the form of imputed generators of the illumination cone. Our procedure does not require expensive "inverse graphics" datasets, and sees no ground truth data of any kind.
Qualitative evaluation suggests the method can erase and restore soft indoor shadows, and can "steer" light around a scene. We offer a summary quantitative evaluation of the method with a novel application of the FID. An extension of the FID allows per-generated-image evaluation. Furthermore, we offer qualitative evaluation with a user study, and show that our method produces images that can successfully be used for data augmentation.
△ Less
Submitted 8 December, 2021;
originally announced December 2021.
-
Deep 3D Convolutional Neural Network for Automated Lung Cancer Diagnosis
Authors:
Sumita Mishra,
Naresh Kumar Chaudhary,
Pallavi Asthana,
Anil Kumar
Abstract:
Computer Aided Diagnosis has emerged as an indispensible technique for validating the opinion of radiologists in CT interpretation. This paper presents a deep 3D Convolutional Neural Network (CNN) architecture for automated CT scan-based lung cancer detection system. It utilizes three dimensional spatial information to learn highly discriminative 3 dimensional features instead of 2D features like…
▽ More
Computer Aided Diagnosis has emerged as an indispensible technique for validating the opinion of radiologists in CT interpretation. This paper presents a deep 3D Convolutional Neural Network (CNN) architecture for automated CT scan-based lung cancer detection system. It utilizes three dimensional spatial information to learn highly discriminative 3 dimensional features instead of 2D features like texture or geometric shape whick need to be generated manually. The proposed deep learning method automatically extracts the 3D features on the basis of spatio-temporal statistics.The developed model is end-to-end and is able to predict malignancy of each voxel for given input scan. Simulation results demonstrate the effectiveness of proposed 3D CNN network for classification of lung nodule in-spite of limited computational capabilities.
△ Less
Submitted 4 May, 2019;
originally announced June 2019.
-
Novel Hard Link-Switching Scheme using Pre-Scanning for Indoor VLC Networks
Authors:
Sumita Mishra,
Sachin Kumar,
Shivani Singh,
Pallavi Asthana,
Nidhi Mathur
Abstract:
VLC is attracting a lot of attention as an emerging potential technology for deployment in next generation indoor wireless networks. Use of efficient link switching scheme among VLC access points is critical in indoor environment to provide seamless connectivity to mobile users. This paper presents a novel position prediction link switching scheme for indoor visible light communication systems. Th…
▽ More
VLC is attracting a lot of attention as an emerging potential technology for deployment in next generation indoor wireless networks. Use of efficient link switching scheme among VLC access points is critical in indoor environment to provide seamless connectivity to mobile users. This paper presents a novel position prediction link switching scheme for indoor visible light communication systems. The method exploits the fact that indoor scenario (light fixtures/furniture) mostly remains unchanged, therefore, this information can be stored at the coordinator end. Thus, the user is not required to search for the best transmitter when RSS is reduced to a threshold value as in conventional methods which cause substantial delay in switching process. Simulation results show that the proposed scheme for indoor VLC System achieves the dual purpose of adequate illumination and mobility to user in considered indoor scenario.
△ Less
Submitted 21 April, 2018;
originally announced May 2018.
-
Automated Detection of Acute Leukemia using K-mean Clustering Algorithm
Authors:
Sachin Kumar,
Sumita Mishra,
Pallavi Asthana,
Pragya
Abstract:
Leukemia is a hematologic cancer which develops in blood tissue and triggers rapid production of immature and abnormal shaped white blood cells. Based on statistics it is found that the leukemia is one of the leading causes of death in men and women alike. Microscopic examination of blood sample or bone marrow smear is the most effective technique for diagnosis of leukemia. Pathologists analyze mi…
▽ More
Leukemia is a hematologic cancer which develops in blood tissue and triggers rapid production of immature and abnormal shaped white blood cells. Based on statistics it is found that the leukemia is one of the leading causes of death in men and women alike. Microscopic examination of blood sample or bone marrow smear is the most effective technique for diagnosis of leukemia. Pathologists analyze microscopic samples to make diagnostic assessments on the basis of characteristic cell features. Recently, computerized methods for cancer detection have been explored towards minimizing human intervention and providing accurate clinical information. This paper presents an algorithm for automated image based acute leukemia detection systems. The method implemented uses basic enhancement, morphology, filtering and segmenting technique to extract region of interest using k-means clustering algorithm. The proposed algorithm achieved an accuracy of 92.8% and is tested with Nearest Neighbor (KNN) and Naive Bayes Classifier on the data-set of 60 samples.
△ Less
Submitted 6 March, 2018;
originally announced March 2018.
