Coqueta Studio - local video segmentation

MediaPipe Tasks Vision 0.10.21, Apache License 2.0 (LICENSE.txt).
Source: https://github.com/google-ai-edge/mediapipe/tree/v0.10.21
Pinned runtime: https://www.npmjs.com/package/@mediapipe/tasks-vision/v/0.10.21

Google MediaPipe models:
Selfie segmenter float16 v1:
https://storage.googleapis.com/mediapipe-models/image_segmenter/selfie_segmenter/float16/1/selfie_segmenter.tflite
Magic Touch float32 v1:
https://storage.googleapis.com/mediapipe-models/interactive_segmenter/magic_touch/float32/1/magic_touch.tflite
EfficientDet-Lite2 int8, local snapshot 2026-09-24 (person detector):
https://storage.googleapis.com/mediapipe-models/object_detector/efficientdet_lite2/int8/latest/efficientdet_lite2.tflite
SHA256: B3F50554CB0EA559E90328845F7D9BA4D13C8BFF372914D24E06BC8BB72FA896
Local file: efficientdet-lite2-int8-v1.tflite (v1 is the local snapshot name).
Model information: https://ai.google.dev/edge/mediapipe/solutions/vision/object_detector
Upstream EfficientDet code: https://github.com/google/automl/tree/master/efficientdet (Apache-2.0).

Selfie Multiclass 256x256 float32, Apache-2.0, local snapshot 2026-09-23:
https://storage.googleapis.com/mediapipe-models/image_segmenter/selfie_multiclass_256x256/float32/latest/selfie_multiclass_256x256.tflite
SHA256: C6748B1253A99067EF71F7E26CA71096CD449BAEFA8F101900EA23016507E0E0
Model card: https://storage.googleapis.com/mediapipe-assets/Model%20Card%20Multiclass%20Segmentation.pdf
Model information and applicable terms:
https://ai.google.dev/edge/mediapipe/solutions/vision/image_segmenter
https://ai.google.dev/edge/mediapipe/solutions/vision/interactive_segmenter
https://ai.google.dev/edge/mediapipe/solutions/vision/interactive_segmenter#models

MediaBunny 1.56.2: see ../video/NOTICE.txt.
GIF export: gifenc 1.0.3, MIT, Copyright (c) 2017 Matt DesLauriers.
Source: https://github.com/mattdesl/gifenc
Pinned distribution: https://www.npmjs.com/package/gifenc/v/1.0.3
Local module: ../video/gifenc-1.0.3.mjs
Full license: ../video/gifenc-LICENSE.txt
No commercial API is called. Model results are approximate.