Waymo’s Post

Excited to share my recent publication in the Machine Vision and Applications Journal! 🎉 Our paper, "RGB-LiDAR fusion for accurate 2D and 3D object detection," is a culmination of our efforts to benchmark and analyze fusion techniques across 2D and 3D detection landscapes, especially for enhancing perception in autonomous driving systems. Special appreciation to my supervisor, Prof. Robert Laganiere, and my co-authors, Morteza Mousa Pasandi and Tianran Liu. Find the full paper here: https://lnkd.in/dQ8-ihAb #computervision #objectdetection #sensorfusion #deeplearning #research

How do you make a self-driving car stop? This was the question my daughter (8) asked me this morning on our walk to school. I'd been telling my kids that I spent yesterday at Zenzic's CAM Innovator's day talking about (amongst other things) self-driving cars. It's a simple question, right? But I didn't know the answer. She wasn't asking about "brake by wire" or about the complex interaction between AI, sensors and edge processors. She wanted to know how we get a self-driving car to pull over because we've just seen a sign for ice cream up ahead. Or, put another way - How do we get a self-driving car to pull over so that we can watch the sunset? FYI: I spend a lot of time talking to customers about use-cases and product development. If you'd like to talk to me about your product or perhaps you'd like to know how Geospatial technology might benefit you (or maybe ask what it does) then give me a shout - I don't bite. #selfdrivingcars #connectedcars #geospatial #consultantforhire

💡Imagine how computer vision models would improve if each pixel in your images or videos could tell a story. 🤝 With the power of semantic segmentation, assign a specific label to every pixel of your training data. The potential applications are limitless, from autonomous vehicles to healthcare diagnostics. Now, with the introduction of V7's new semantic masks, this potential can be fully unlocked! ➡️ Learn more about masks and their wide range of applications: https://lnkd.in/eMRVrchi

Greetings LinkedIn community! 👋 Excited to share our latest blog, delving into “An Insight into Optical Flow and Its Application using Lucas-Kanade Algorithm.” 🌐💡 Explore the dynamic world of Optical Flow and its application using the Lucas-Kanade Algorithm in our latest blog by Chinmay Gupta . Uncover the power of tracking object movement in real-world scenarios and its significance in fields like Autonomous Driving and Medical Research. Read more about it here: https://lnkd.in/g7BsTsFw #SPARK4AI #OpticalFlow #LucasKanade #ComputerVision #TechInsights 🌐💡

⚡Anyverse and Sony Semiconductor are collaborating to ensure sensor and AI model seamless integration to accelerate ADAS deployment Thanks to Anyverse synthetic data solution incorporating Sony Image Sensor Models, perception teams using or planning on using those sensors can now generate highly accurate synthetic datasets and physically-based sensor simulation to evaluate and test their Advanced Driver Assistance System (ADAS) and Automated Driving AI models. 🟢Learn more https://lnkd.in/dDZrEWun #Automotive #ADAS #AutonomousDriving #Sony #SensorSimulation #ImageSensors #MachineVision

🧪 New Machine Learning Research: Semantic Segmentation for Autonomous Driving Discover the Context-Aware Interaction Network (CAINet). It revolutionizes RGB-T semantic segmentation for enhanced driving scene understanding. ➡️ Dive into details in the carousel post! - - - Found this valuable? Hit follow Label Your Data🔔 for more posts like this one. #LabelYourData #SemanticSegmentation #MLresearch #AIresearch

AVCC and MLCommons Reveal Industry Benchmark for Automotive Perception with Cognata Synthetic Dataset! Cognata is proud to announce our significant contribution to the groundbreaking MLPerf Automotive Benchmark Proof-of-Concept (POC) released by MLCommons and the Autonomous Vehicle Computing Consortium, Inc. (AVCC). This release marks a crucial milestone in developing a comprehensive benchmark suite for vehicle AI systems. Check the video: running the SSD-ResNet50 Object Detection model at 8 megapixel image input. This model has been trained using a large synthetic dataset from Cognata, rendered with 8 megapixel resolution. With this sophisticated training regimen, the SSD-ResNet50 model excels in identifying specific road objects which are not just recognized but also accurately labeled and highlighted within precise bounding boxes. Read the full press release: https://lnkd.in/gqGuAVnn Alon Atsmon Danny Atsmon Shay Rootman Nofar Assis Gadi Sznajder Omri Reftov Arm Bosch NVIDIA ABTF Automotive Business Task Force cKnowledge Marvell Technology QUALCOMM TECHNOLOGIES INTERNATIONAL, LTD. #syntheticdata #autonomousdriving #digitaltwin #photorealistic #machinelearning #autonomousvehicle #automotiveindustry #AutomotiveInnovation #VehicleSimulation

AI is a very powerful solution for complex tasks in an automated vehicle. The safety and system engineering challenges for vehicles with AI components are however still an open question. During the online session on July 11th, experts @Jan-Pieter Paardekooper (TNO) and @Riccardo Dona (European Commission JRC) illuminate the underlying issues and discuss potential solutions. A great opportunity to gain insights into this area! 🔍 🚗 Join ->  July 11th | Safe automated driving | Meet the expert (tno.nl) 🚗 https://lnkd.in/eMYf7PAs #automateddriving

I'm excited to be a speaker at Auto.AI in Berlin, September 24-26. I will be talking about the importance of accurate synthetic data for camera perception training & testing. With a specific focus on how we manage to generate such data with our physics-based approach, and a few edge cases highlights. #AutoAI , #OSS5, #adas, #AutonomousDriving, #Ansys, #simulation

🕛 TODAY at #INCOSEIS 🍀 'AI-Enhanced Autonomous Formation Flying' 📅July 4, 14:15-14:55 | Wicklow Hall 2B This session presented by #Ansys Fellow Bernard Dion, introduces an Autonomy Model-Based Systems Engineering (MBSE) Framework, heavily reliant on #simulation, for developing and validating mission and safety-critical applications, including AI/ML-based constituents within a safety-critical function implemented in a model-based environment. This framework enables users to build digital models, covering mission and vehicle behavior, and lays down the foundations of a digital training and validation environment for autonomous systems, that can provide early and accurate feedback to autonomous systems developers. Furthermore, this framework aims at complying with emerging AI-based safety standards such as the future SAE ARP6983. #systemsengineering #SystemsEngineer #AnsysMBSE #AI #MBSE #SystemArchitecture #AutonomousVehicle #AutonomousVehicleTechnology #SystemSecurity