EPISODE · Nov 10, 2025 · 32 MIN
Computer Vision | Practise: How Computers See in the Real World
from Big Ideas Only · host Montanus
In this episode of Big Ideas Only, host Mikkel Svold explores how computers “see” with Andreas Møgelmose (Associate Professor of AI, Aalborg University; Visual Analysis & Perception Lab). We unpack what computer vision is, where it already works at scale, what’s still hard, and the real-world trade-offs around privacy and surveillance - from self-driving cars and robots to hospital X-rays and trash sorting.In this episode, you’ll learn about:What computer vision really is: turning camera input into understanding and actionWhen vision alone is enough, and when you need lidar, radar or time-of-flight sensorsThe biggest driver: industrial automationHow automated triage of X-rays can cut ER waiting times with a doctor reviewing the final resultWhy the classic “who should the car hit?” dilemma misses how real autonomy works3D understanding with stereo cameras and other depth-sensing methodsWhy sorting messy, mixed real-world waste remains one of the hardest vision challengesHumanoid robots — what already works and what’s still far from realityWhere research is headed: from fine-grained recognition to explainability and machine unlearningOn-device versus cloud processing, and how that choice shapes privacy riskEpisode Content 00:01 Why it matters that computers can “see”02:04 When vision alone is enough — and when it isn’t04:40 Healthcare in practice: automated X-ray checks for faster casts and shorter ER waits05:39 Accuracy, human oversight, and how every case gets double-checked in morning rounds07:20 The trolley-problem myth: how real autonomous systems minimize risk instead of choosing victims12:32 Choosing the right approach: classification versus 3D navigation13:36 Getting depth: stereo vision, lidar, radar, and time-of-flight sensors16:01 Why sorting mixed, messy waste is still one of the hardest vision problems18:03 Humanoid robots: balance, stairs, and why sight is the foundation for movement19:21 Status check: “solved” in some areas, far from it in others20:40 Privacy and ethics: on-device versus cloud processing, and who controls the data27:37 What’s still missing: fine-grained recognition, explainability, and machine unlearning32:28 Current projects: pre-anesthesia screening, color detection in video, and robust segmentation33:32 Outro and teaser for a deeper theoretical dive next episodeThis podcast is produced by Montanus.
What this episode covers
In this episode of Big Ideas Only, host Mikkel Svold explores how computers “see” with Andreas Møgelmose (Associate Professor of AI, Aalborg University; Visual Analysis & Perception Lab). We unpack what computer vision is, where it already works at scale, what’s still hard, and the real-world trade-offs around privacy and surveillance - from self-driving cars and robots to hospital X-rays and trash sorting.In this episode, you’ll learn about:What computer vision really is: turning camera input into understanding and actionWhen vision alone is enough, and when you need lidar, radar or time-of-flight sensorsThe biggest driver: industrial automationHow automated triage of X-rays can cut ER waiting times with a doctor reviewing the final resultWhy the classic “who should the car hit?” dilemma misses how real autonomy works3D understanding with stereo cameras and other depth-sensing methodsWhy sorting messy, mixed real-world waste remains one of the hardest vision challengesHumanoid robots — what already works and what’s still far from realityWhere research is headed: from fine-grained recognition to explainability and machine unlearningOn-device versus cloud processing, and how that choice shapes privacy riskEpisode Content 00:01 Why it matters that computers can “see”02:04 When vision alone is enough — and when it isn’t04:40 Healthcare in practice: automated X-ray checks for faster casts and shorter ER waits05:39 Accuracy, human oversight, and how every case gets double-checked in morning rounds07:20 The trolley-problem myth: how real autonomous systems minimize risk instead of choosing victims12:32 Choosing the right approach: classification versus 3D navigation13:36 Getting depth: stereo vision, lidar, radar, and time-of-flight sensors16:01 Why sorting mixed, messy waste is still one of the hardest vision problems18:03 Humanoid robots: balance, stairs, and why sight is the foundation for movement19:21 Status check: “solved” in some areas, far from it in others20:40 Privacy and ethics: on-device versus cloud processing, and who controls the data27:37 What’s still missing: fine-grained recognition, explainability, and machine unlearning32:28 Current projects: pre-anesthesia screening, color detection in video, and robust segmentation33:32 Outro and teaser for a deeper theoretical dive next episodeThis podcast is produced by Montanus.
NOW PLAYING
Computer Vision | Practise: How Computers See in the Real World
No transcript for this episode yet
Similar Episodes
No similar episodes found.