video1.0<iframe src="https://www.loom.com/embed/e44c35ec0c944b599ca8e2982694145c" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/sessions/thumbnails/e44c35ec0c944b599ca8e2982694145c-36957197704cde07.gif246.038Hi everyone, I am Saran Saini, and I am demonstrating my DSAI Lab project on navigation assistance for visually impaired people in an indoor space. My pipeline takes an RGB frame and runs YOLO object detection and a depth map in parallel, then combines them into depth maps with bounding boxes. From that, I compute hazard zones, including a walkable zone of 40 percent and later zones of 30 percent, and select the layer with the lowest risk. I generate a navigation command and pass it to a TTS module for audio guidance. I deployed the app on HuggingFace, and you can analyze your own images by clicking Analyze Environment.