FrontDepth View vs Vector Search Bench

FrontDepth View provides real-time visualization of both TrueDepth camera data and front camera imagery. You can switch between the front camera view, the TrueDepth depth view, or a combined mode that merges both sources to enhance environmental awareness and depth perception. * FrontDepth modes (Options 3 and 4) are particularly effective in low-light or dark environments, helping to reveal surrounding structures and spatial depth. The visible depth range can be adjusted using the slider at the bottom of the screen. * Combined modes (Options 2 and 4) blend front camera imagery with TrueDepth data to enhance object contours and improve depth detail. * In Combined mode, tap the subview to enlarge it for a closer inspection. Tap again to return to the split or combined layout. * 3D Point Cloud View allows you to freely explore your surroundings in an interactive 3D perspective by panning the screen. * The center-point depth is displayed in real time in centimeters. You can capture screenshots at any time by tapping the camera icon, and view your saved images by tapping the photo gallery icon. If you want to view back depth image through LiDAR and back camera, please check our app "LiDAR View". If you want to automatically measure facial features in real time using the TrueDepth camera, please check our app “Measure Face”. If you have any suggestions, feedback, or feature requests, feel free to contact us at hereyouare1981@gmail.com Thank you for using FrontDepth View!

Vector Search Bench is a high-signal, no-nonsense benchmark suite for vector similarity search on iPhone. Measure how fast and stable your device can score embeddings using the CPU (Accelerate/vDSP) versus the Apple Neural Engine (Core ML)—with real metrics you can trust. What you can benchmark Exact Top-K vector search (dot-product similarity) CPU path: Accelerate / vDSP optimized scoring NPU path: Core ML Dense inference tuned for Neural Engine scheduling Query-by-query execution with repeatable timing Intricate performance metrics Latency distribution: median, p95 Throughput: QPS Recall@K vs CPU reference (sanity check) Memory footprint and dataset sizing knobs Built for real-world AI systems