Sorry, you need to enable JavaScript to visit this website.

Photonic circuits pave the way to ultrafast computing and real-time inference of applications with paramount importance, such as imaging flow cytometry (IFC). However, current implementations exhibit inherent restrictions that consequently diminish the neural networks (NN) complexity that can be supported. Thus, NN compression mechanisms are deemed critical for the efficient deployment of such demanding tasks.

Image compression is a key problem in this age of information explosion. With the help of machine learning, recent studies have shown that learning-based image compression methods tend to surpass traditional codecs. Image compression can be split into three steps: transform, quantization, and entropy estimation.

Object pose estimation remains an open and important task for autonomous systems, allowing them to perceive and interact with the surrounding environment. To this end, this paper proposes a 3D object pose estimation method that is suitable for execution on embedded systems. Specifically, a novel multi-task objective function is proposed, in order to train a Convolutional Neural Network (CNN) to extract pose-related features from RGB images, which are subsequently utilized in a Nearest-Neighbor (NN) search-based post-processing step to obtain the final 3D object poses.

In this paper, we propose a framework for 3D human pose estimation with a single 360° camera mounted on the user's wrist. Perceiving a 3D human pose with such a simple setting has remarkable potential for various applications (e.g., daily-living activity monitoring, motion analysis for sports enhancement). However, no existing work has tackled this task due to the difficulty of estimating a human pose from a single camera image in which only a part of the human body is captured and the lack of training data.

We address 3D human pose and shape estimations from multi-view images. We use the SMPL body model, and regress the model parameters that best fit the shape and pose. To solve for the parameters, we first compute 3D joint positions from 2D joint estimations on images by using a linear algebraic triangulation. Then, we fit the 3D parametric body model to the 3D joints while imposing a bone orientation constraint between the 3D model and the corresponding body parts detected in the images.

At present, the performance of the end-to-end stereo matching networks based on CNN greatly exceed the traditional stereo matching networks, but the accuracy in those ill-posed regions like foreground areas is still not optimistic. In this paper, we propose a novel design to improve the prediction performance of disparity in foreground. First, a multi-scale pyramid aggregation module with hourglass-like structure is designed to effectively utilize the aggregation information of different scales.

A singular problem that mars the wide applicability of machine learning (ML) models is the lack of generalizability and interpretability. The ML community is increasingly working on bridging this gap. Prominent among them are methods that study causal significance of features, with techniques such as Average Causal Effect (ACE). In this paper, our objective is to utilize the causal analysis framework to measure the significance level of the features in binary classification task.

Autonomous Vehicles promise to transport people in a safer, accessible, and even efficient way. Nowadays, real-world autonomous vehicles are build by large teams from big companies with a tremendous amount of engineering effort. Deep Reinforcement Learning can be used instead, without domain experts, to learn end-to-end driving policies. Here, we combine Curriculum Learning with deep reinforcement learning, in order to learn without any prior domain knowledge, an end-to-end competitive driving policy for the CARLA autonomous driving simulator.