Processing 3D images has many use cases. For example, to improve autonomous car driving, to enable digital conversions of old factory buildings, to enable augmented reality solutions for medical surgeries, etc. The size of the point cloud, the number of points, sparse and irregular point cloud, and the adverse impact of the light reflections, (partial) occlusions, etc., make it difficult for engineers to process point clouds. 3D Point Cloud image processing is increasingly used to solve Industry 4.0 use cases to help architects, builders and product managers. I will share some of the innovations that are helping the progress of 3D point cloud processing. I will share the practical implementation issues we faced while developing deep learning models to make sense of 3D Point Clouds.

The frameworks for deep learning provides the critical blocks for designing, training and validating deep networks. They offer a high level of abstraction with the flexible and simple programming interface to diverse community of deep learning practitioners. Undoubtedly, Tensorflow is emerged as the industry's most popular framework adopted by many software giants due to its highly scalable and flexible system architecture. When its comes to easy programming interface and debugging, it is still a pain due to its statistic graph's construction approach. The tools like tf Debugger and Tensorboard are especially introduced to make user's life simple to debug and visualize the state of learning. In this deep dive we will try to understand deep learning concepts applied in computer vision and image orocessing.When it comes to the myriad of applications with images, these tools helps the user in understanding the state of a learning algorithm. We will examine a simple deep learning based image recognition program, breaks it and learn to fix it effectively with the help of tools like tf debugger and Tensorboard. We will then try to understand the sophisticated concepts of tuning the deep network by visualizing the effects on Tensorboard.We will also explore the much more complex project like image segmentation for real world data and image domain transfer with adversarial networks with Tensorboard. We will learn how these handy tools can really make a big difference in deciphering the complexities of the algorithm and remove the "Dont know what to do next ! " situation in DL practitioner's life.

Nowadays, there are various trends on busting training and execution of deep neural networks using hardware acceleration. Thus, it is worthwhile to take a deeper look at using FPGA to burst the execution of DNNs to new spheres. Using pre-trained models and adapt them via transfer learning to specific problems pushes FPGA at least for execution of DNNs in a different league. This talk will show how to use FPGAs in specific scenarios for quality checks in factories or for security surveillance. This talk will provide some more insights in how to use FPGAs and accelerate the execution of AI model

Image recognition is an essential part of autonomous driving technology. Cars have to recognise a multitude of items in a front camera scene. Deep learning networks are the state-of-the-art modelling approach to take decisions based on the camera image feed. Unfortunately, research has shown that universal perturbations on this image feed can be designed such as to corrupt the networks’ decisions. This fact has strong implications on the security and safety of autonomous cars today. This deep dive explains how perturbations work and if and how they can be detected in the data. The talk includes a live demonstration where a perturbation is constructed from data and applied to a self-driving car’s street scene.

With the growing amount of products and information, with significant rise number of users, it becomes increasingly important for companies to search, map and provide them with the relevant chunk of information or products according to their preferences and tastes. Let's talk about deep learning approach in recommender systems, which is gaining more and more popularity, their advantages and disadvantages, and the specific scenarios in which they are most effective.

Finding anomalies is only one part of a comprehensive repair/maintenance solution. Not all anomalies are problems and not all problems need to be fixed. Understanding the context behind a sensor reading — where anomalies come from, what they mean, and what needs to be done about them — requires extracting the intent-data from other sources, be they historical service orders, OEM manuals, or human heuristics. This gives symptom and resolution information that guides the process from complaint to correction. This presentation covers our AI models that enable intent discovery in one of the noisiest domains: Automotive. Learn how we made sense of data coming from 70,000+ different vehicle makes and models.

Human languages are complex, diverse and riddled with exceptions – translating between different languages is therefore a highly challenging technical problem. Deep learning approaches have proved powerful in modelling the intricacies of language, and have surpassed all statistics-based methods for automated translation. This session begins with an introduction to the problem of machine translation and discusses the two dominant neural architectures for solving it – recurrent neural networks and transformers. A practical overview of the workflow involved in training, optimising and adapting a competitive neural machine translation system is provided. Attendees will gain an understanding of the internal workings and capabilities of state-of-the-art systems for automatic translation, as well as an appreciation of the key challenges and open problems in the field.