LooperRobotics

The wait is over. Meet Insight 9👁️ Native Spatial AI. 24g IMU. All packed in a Premium Aluminum Alloy Case. We are redefining vision for Embodied AI. Coming exclusively to Kickstarter. #SpatialAI #Robotics #Kickstarter #TechLeak #EmbodiedAI #TechNews #TechTrends

Get a sneak peek at Insight 7 in our official teaser video. 🎬 Designed for compact mobile robotic platforms, Insight 7 features a 70 mm standard stereo baseline and an ideal range of 0.15 m to 15 m, optimized for mid-range perception. With a more compact form factor, Insight 7 inherits the same onboard VIO, 10 TOPS AI acceleration engine, and ROS integration capabilities as Insight 9. Equipped with Dual Global Shutter sensors, it delivers synchronized visual, spatial, and motion data out of the box. Coming soon. 🦾 #SpatialAI #Robotics #ComputerVision #EmbodiedAI #ROS #VSLAM

High-quality robot learning starts with high-quality data. This sequence demonstrates a standardized Embodied AI data collection and real-time verification process covering articulated object manipulation, deformable object manipulation, and fine-grained manipulation. One of the biggest engineering challenges is reducing offline post-processing caused by trajectory drift, inaccurate frontend VIO, and time synchronization errors. The Looper Embodied Data Collection Suite synchronously captures RGB, depth, IMU, and real-time VIO poses, supporting a streamlined workflow from collection to delivery. High-quality data production is the essential infrastructure for Embodied AI. #Robotics #EmbodiedAI #MachineLearning #ComputerVision #PhysicalAI #RobotLearning #DataCollection #UMI #EgoData

Gripper-mounted Ego + UMI data collection. This demo explores a new workflow using a head-mounted Insight 9 together with Insight 7 mounted on the grippers. ▪️ Object picking ▪️ Cloth folding Synchronously capturing: ▪️ First-person observations ▪️ End-effector trajectories ▪️ Real-time VIO Not just capturing data—it enables real-time verification during collection. P.S. In the current demo, gripper opening/closing angle is not yet synced. We'll update this in our next test. Stay tuned! #EmbodiedAI #RobotLearning #Robotics #EgoData #UMI #VIO #PhysicalAI #ComputerVision

🚀 Insight 9 running on NVIDIA Jetson Orin NX. Standard USB Camera Mode Tutorial : ✅ Insight Viewer installation ✅ Automatic device detection ✅ RGB & Stereo grayscale stream ✅ Real-time depth map ✅ Real-time 3D motion (VIO) trajectory No ROS setup. Uses standard USB camera drivers. Just connect and start streaming. #JetsonOrinNX #JetsonOrin #NVIDIA #EdgeAI #ComputerVision #Robotics #EmbodiedAI #DepthCamera

Meet the next member of the Insight family: Insight 7. Designed for compact mobile robotic platforms, Insight 7 features a 70 mm standard stereo baseline and an ideal range of 0.15 m to 15 m, optimized for mid-range perception. With a more compact form factor, Insight 7 inherits the same onboard VIO, AI acceleration, and ROS integration capabilities as Insight 9, delivering synchronized visual, spatial, and motion data out of the box. Coming soon. 🦾 #LooperRobotics #SpatialAI #Robotics #ComputerVision #EmbodiedAI #ROS

This clip demonstrates an outer loop running in the physical world. Insight 9 + Insight 7 capture synchronized multi-modal data (vision + VIO trajectories) with real-time on-device verification — no heavy post-processing required. Clean, persistent artifacts are generated during task execution. These artifacts compound across sessions and directly feed better inner loops for autonomous robots. This is embodied Loop Engineering. #LoopEngineering #EmbodiedAI #VIO #Robotics #PhysicalAI

Native ROS deployment on NVIDIA Jetson Orin NX. ▪️ Jetson Orin NX zero-configuration setup ▪️ ROS integration ▪️ Live 4-channel visualization & Depth map ▪️ Real-time VIO tracking Insight 9 running directly on NVIDIA Jetson Orin NX through a ROS workflow. Part 2: UVC mode coming soon. #JetsonOrin #JetsonOrinNX #NVIDIA #ROS #Robotics #EmbodiedAI #EdgeAI #ComputerVision

@SanSanychUA Exactly! Our VIO pipeline inherently fuses the IMU data right out of the box. This seamlessly integrated VIO data stream is part of the raw output from Insight 9, making it readily available for everyone to use directly for their applications

@LooperRobotics Pure VIO holding near-zero drift through the ICRA lobby with glass walls and foot traffic is genuinely impressive — most systems fold at the first reflective surface. The colorized point cloud in real time is the part that catches attention. Are you fusing any additional IMU pre-integration beyond the default VIO pipeline, or is this raw Insight 9 output?

Insight 9 + Insight 7 for Ego + UMI Data Collection. 1× Insight 9 (Head-mounted) 2× Insight 7 (Wrist-mounted) Synchronously capturing: ▪️ First-person observations ▪️ Hand motion trajectories ▪️ Real-time VIO poses Task: Chinese Kung Fu (Tai Chi). All computed on-device and available in real time. Verify results during collection. Less post-processing. Faster iteration. #EmbodiedAI #EgoData #UMI #PhysicalAI #VIO #Robotics #SpatialAI

@alexpdiggs Thanks! 🙌 That’s Insight 9, the world’s 1st Autonomous Spatial AI Camera. It gives robots ultra-smart "eyes" with high-precision spatial perception. 🔗 Visit our official website for details!

@LooperRobotics looooks sweet. what is it?

🚀 Exciting progress from the lab! Our mass production is in the final testing and calibration phase, and the first batch of Insight 9 units is looking fantastic. 👀 📦 To our Kickstarter Backers: Your Backer Surveys are officially OUT! Please check your Kickstarter inbox and submit your shipping info ASAP (a valid phone number is absolutely required for international delivery). ⚠️ Quick expectation check: We are collecting your addresses IN ADVANCE. We are not shipping boxes out today. By getting your info sorted now, we ensure zero delays the exact moment our mass production and final testing are 100% complete! 🌟 Missed the Kickstarter? Official website pre-orders are now open! Secure your unit to join our upcoming shipping waves: shorturl.at/iA6nz #Robotics #SpatialAI #EmbodiedAI #LooperRobotics

Insight 9 + Insight 7 for Ego + UMI Data Collection. 1× Insight 9 (Head-mounted) 2× Insight 7 (Wrist-mounted) Synchronously capturing: ▪️ First-person observations ▪️ Hand motion trajectories ▪️ Real-time VIO poses Tasks: Folding clothes. Pouring water. Object placement. Writing. Capturing the full process from perception to action in a single workflow. #EmbodiedAI #EgoData #UMI #PhysicalAI #VIO #Robotics #SpatialAI

SLAM test in the lobby of the ICRA 2026 venue. Wide-open spaces. Glass reflections. Continuous foot traffic. Insight 9 maintained stable tracking. Running pure VIO with NO loop closure, it achieved near-zero drift while generating a colorful 3D point cloud in real time. 💡Note: All VSLAM & depth compute runs on-board & streams via ROS! Laptop is ONLY for registration & visualization, no extra ICP or BA. #ICRA2026 #SLAM #Robotics #PointCloud #ROS #EmbodiedAI

Exploring real-time data collection for Embodied AI. Object picking. Object placement. Surface wiping. Insight 9 records synchronized first-person observations and VIO trajectories, helping streamline Ego & UMI data collection workflows. #EmbodiedAI #EgoData #UMI #PhysicalAI #VIO

UMI & Ego data without MoCap! Insight 9 handheld VIO test for Embodied AI: ▪️ Pure on-device real-time ▪️ 0 post-processing ▪️ No loop closure ⏱️ 16.1s | 📏 2.86m | 🎯 3.2mm endpoint error #VIO #Robotics #PhysicalAI

At ICRA 2026, Insight 9 underwent aggressive motion testing. 🦾 Continuous shaking. Rapid movements. Unpredictable trajectories. Insight 9 maintained stable visual-inertial tracking throughout the test. #ICRA2026 #SLAM #VIO #Robotics

⚡️ macOS Deployment in Under 60s. Deploy Insight 9 natively on macOS with a streamlined Plug & Play workflow. Built-in Linux edge node & 10 TOPS onboard compute power real-time spatial data streaming on macOS. #ROS2 #Robotics #macOS

No LiDAR. Just Vision. 🤖 Outdoor slopes & smooth indoor floors challenge quadruped localization. We adopt Insight 9 + TinyNav visual odometry for stable autonomous navigation. Outdoor ➡️ Indoor transition. #Unitree #Robotics #VSLAM

@techietaro Great idea! Really looking forward to it!

@LooperRobotics What if I train the world model to predict the light shift before the sensor catches up? Predictive perception eliminates the depth drop entirely...

Insight 9 FOV & HDR Challenge! 💡 Sudden light transitions can cause exposure lag & depth data loss in Embodied AI vision, while a narrow FOV limits perception. #EmbodiedAI #Robotics #ComputerVision #VSLAM