Wednesday, April 1, 2020
ML solutions for Live Perception (TF Dev Summit '20)
Live Perception is a challenging scenario enabled by on-device ML, in real-time, and with low-latency. In this talk you will learn about best practices and how to use TensorFlow Lite and TF.js to enable Live Perception solutions by diving into the technical details of 3 unique ML solutions: Face Meshes, Hand Poses and Object cuboids. Speaker: Matthias Grundmann - Research Scientist Resources: BlazeFace → https://goo.gle/2Tves6j GitHub Face Detection (GPU) → https://goo.gle/39wK8NY GitHub BlazeFace → https://goo.gle/39uCX9f Octi → https://goo.gle/2PT1Bs2 Model Card - BlazeFace Face Detection → https://goo.gle/2TFVKHQ On-Device, Real-Time Hand Tracking with MediaPipe blog → https://goo.gle/3auULRq Face Mesh Demo → https://goo.gle/39uDFDr GitHub MediaPipe → https://goo.gle/38un7Kf Watch all TensorFlow Dev Summit 2020 sessions → https://goo.gle/TFDS20 Subscribe to the TensorFlow YouTube channel → https://goo.gle/TensorFlow
Subscribe to:
Post Comments (Atom)
Most watched
-
Quick answer: Qwen 3 32B needs about 22.7 GB of VRAM at Q4_K_M with a 4k-token context — a used 24 GB card like the RTX 3090 is the minimu...
-
Using GPUs in TensorFlow, TensorBoard in notebooks, finding new datasets, & more! (#AskTensorFlow) [Collection] In a special live ep...
-
Inside TensorFlow: Summaries and TensorBoard [Collection] Take an inside look into the TensorFlow team’s own internal training sessions-...
-
Quick answer: For an 8B model at Q4, local electricity costs roughly $0.05 per million output tokens — against about $0.53 per million ble...
-
Quick answer: For machine-learning work, buy an external SSD over a spinning hard drive, pick USB 3.2 Gen 2 (10 Gbps) or faster, and size f...
-
Quick answer: DeepSeek V4 Flash is a Mixture-of-Experts model: all 284 billion parameters must be stored, so combined RAM plus VRAM — not ...
-
Quick answer: Yes, for most budget builders: a used RTX 3090 is still the cheapest route to 24 GB of VRAM with full CUDA support. Capacity...
-
Hey everyone This is Ujjwal kapil B.tech 2nd year Ai and Ml In this video we covered one of the most important question "COLLEGE KAB O...
-
Quick answer: The cheapest path to serious local-AI VRAM is not one new flagship card — it is two used ones. A single RTX 5090 costs $1,999...
-
Quick answer: FLUX.1 dev needs about 10 GB of VRAM at Q4/NF4 quantization for 1024×1024 image generation, so a 12 GB GPU such as the GeFor...
No comments:
Post a Comment