Loading...
Loading...
Found 55 Skills
Search video archives using natural language — find events, objects, actions, and people across recorded video using fusion search (Cosmos Embed1 semantic search + CV attribute search). Use when asked to search for something in video, find actions and events, locate objects and people, or query video archives. For these types of questions, default to this top-level fusion search unless user specifies otherwise. Requires the search profile to be deployed.
Meta Quest Passthrough Camera Access (PCA) for Unity — access the forward-facing RGB cameras on Quest 3 / Quest 3S to feed Computer Vision and Machine Learning pipelines. Use when capturing the passthrough camera image/texture, reading the camera pose and intrinsics, projecting camera pixels into world space via `PassthroughCameraAccess.ViewportPointToRay`, wiring camera frames into ML/CV models, or reasoning about resolution, permissions, vendor tags, and the pinhole/principal-point model. For projecting the image onto a flat world-space surface (frustum-slice quad / image-plane overlay) see references/principal-point-offset.md; for placing 2D ML detections as world-space 3D bounding boxes see references/detection-bounding-boxes.md. Skip if the user only wants to cast the user's POV (use the Media Projection API instead) or is doing screen-space-only overlays.
Apply, configure, and modulate WebGPU post-processing shaders, cinematic color grading, tone mapping, procedural VFX, and on-device vision (object tracking, instance segmentation, pose, person mattes) in framefields. Use when adding blur, grain, LUTs, relighting, curves, greenscreen color keying, subject cutouts, or tracking-driven effects in framefields compositions.
Membangun game interaktif berbasis Augmented Reality (AR) dan deteksi gerakan kamera (Motion & Gesture Tracking) untuk Smart TV, Interactive Flat Panel (IFP), papan tulis digital, atau webcam laptop. Memungkinkan siswa berinteraksi di depan kelas tanpa menyentuh layar fisik.
Segment overlapping visual parts from source images, wireframes, texture atlases, and decals before mesh reconstruction. Use when a mascot/logo/template contains touching or overlapping components and exact structural part masks are needed before contour-to-mesh, UV fitting, or landmark repair.
Turn a single-person or centrally framed single-pet photo plus a square stylized target into a deterministic, verifiable stroke-by-stroke hand-drawing video. Use when the user asks to 把照片变成逐笔手绘视频, 照片临摹动画, 真实画笔跟随, marker/dry-brush drawing process, or a no-reveal drawing video rendered with finite-width strokes rather than video generation or target-image compositing.
Knowledge base on spatial intelligence as the next frontier in AI, based on Fei-Fei Li's insights from her Y Combinator AI Startup School talk. Use this skill when users ask about spatial intelligence concepts, 3D world modeling, the evolution of computer vision from ImageNet to World Labs, AI research strategy and problem selection, or when seeking advice on AI entrepreneurship and founding AI companies. Also trigger when discussing the relationship between vision/spatial understanding and AGI, differentiating generative vs discriminative models in 3D contexts, or exploring the data-algorithm-compute trinity for AI breakthroughs.