🤖 Model fire/v-sekai.mediapipe-labeler Label facial blendshapes and pose keypoints from images or videos. Detect up to 1–100 people and return per-frame JSON a... pose-estimation • face-detection • facial-blendshapes • 218 runs