Recently, several methods have been proposed for estimating the mutual information from sample data using deep neural networks and without the knowing closed form distribution of the data. This class of estimators is referred to as neural mutual information estimators. Although very promising, such techniques have yet to be rigorously bench-marked so as to establish their efficacy, ease of implementation, and stability for capacity estimation which is joint maximization frame-work. In this paper, we compare the different techniques proposed in the literature for estimating capacity and provide a practitioner perspective on their effectiveness. In particular, we study the performance of mutual information neural estimator (MINE), smoothed mutual information lower-bound estimator (SMILE), and directed information neural estimator (DINE) and provide insights on InfoNCE. We evaluated these algorithms in terms of their ability to learn the input distributions that are capacity approaching for the AWGN channel, the optical intensity channel, and peak power-constrained AWGN channel. For both scenarios, we provide insightful comments on various aspects of the training process, such as stability, sensitivity to initialization.
Purpose: To estimate the total lung volume (TLV) from real and synthetic frontal chest radiographs (CXR) on a pixel level using lung thickness maps generated by a U-Net deep learning model. Methods: This retrospective study included 5,959 chest CT scans from two public datasets: the lung nodule analysis 2016 (n=656) and the Radiological Society of North America (RSNA) pulmonary embolism detection challenge 2020 (n=5,303). Additionally, 72 participants were selected from the Klinikum Rechts der Isar dataset (October 2018 to December 2019), each with a corresponding chest radiograph taken within seven days. Synthetic radiographs and lung thickness maps were generated using forward projection of CT scans and their lung segmentations. A U-Net model was trained on synthetic radiographs to predict lung thickness maps and estimate TLV. Model performance was assessed using mean squared error (MSE), Pearson correlation coefficient (r), and two-sided Student's t-distribution. Results: The study included 72 participants (45 male, 27 female, 33 healthy: mean age 62 years [range 34-80]; 39 with chronic obstructive pulmonary disease: mean age 69 years [range 47-91]). TLV predictions showed low error rates ($MSE_{Public-Synthetic}$=0.16 $L^2$, $MSE_{KRI-Synthetic}$=0.20 $L^2$, $MSE_{KRI-Real}$=0.35 $L^2$) and strong correlations with CT-derived reference standard TLV ($n_{Public-Synthetic}$=1,191, r=0.99, P<0.001; $n_{KRI-Synthetic}$=72, r=0.97, P<0.001; $n_{KRI-Real}$=72, r=0.91, P<0.001). The Luna16 test data demonstrated the highest performance, with the lowest mean squared error (MSE = 0.09 $L^2$) and strongest correlation (r = 0.99, P <0.001) for TLV estimation. Conclusion: The U-Net-generated pixel-level lung thickness maps successfully estimated TLV for both synthetic and real radiographs.
Human-robot object handover is a key skill for the future of human-robot collaboration. CORSMAL 2020 Challenge focuses on the perception part of this problem: the robot needs to estimate the filling mass of a container held by a human. Although there are powerful methods in image processing and audio processing individually, answering such a problem requires processing data from multiple sensors together. The appearance of the container, the sound of the filling, and the depth data provide essential information. We propose a multi-modal method to predict three key indicators of the filling mass: filling type, filling level, and container capacity. These indicators are then combined to estimate the filling mass of a container. Our method obtained Top-1 overall performance among all submissions to CORSMAL 2020 Challenge on both public and private subsets while showing no evidence of overfitting. Our source code is publicly available: https://github.com/v-iashin/CORSMAL