10dallasj/studiocast

By 10dallasj · ↓ 0 · ♥ 0 · Updated 2026-02-26

What's in this repo

This page lists the files in 10dallasj/studiocast, including gaze_flx_left.onnx, gaze_flx_right.onnx, gaze_flx_left_fp16.onnx and more.

16 ONNX models (.onnx files)
55 files, including 19 .txt, 16 .json, 16 .onnx, 2 .md, 1 .dat
Last updated 2026-02-26

ONNX files

File	Size	Download
eye_contact/Better Quality (Gaze Correction Cam v0.1.1 FP32)/gaze_flx_left.onnx	—	Download
eye_contact/Better Quality (Gaze Correction Cam v0.1.1 FP32)/gaze_flx_right.onnx	—	Download
eye_contact/Good Quality (Gaze Correction Cam v0.1.1 FP16)/gaze_flx_left_fp16.onnx	—	Download
eye_contact/Good Quality (Gaze Correction Cam v0.1.1 FP16)/gaze_flx_right_fp16.onnx	—	Download
face_detection/Best Quality (YuNet FP32)/model.onnx	—	Download
face_detection/Better Quality (YuNet INT8 BQ)/model.onnx	—	Download
face_detection/Good Quality (YuNet INT8)/model.onnx	—	Download
matting/Best Quality/model.onnx	—	Download
matting/Better Quality/model.onnx	—	Download
matting/Good Quality/model.onnx	—	Download

Showing 10 of 16 .onnx files — see the full file list on the right.

How to use

These ONNX files can be loaded with ONNX Runtime. To inspect a model graph, upload the file to the Netron viewer. Download links point to the original files on Hugging Face.

README

StudioCast open_video model pack templates

This directory contains metadata-only templates for open-source video model packs.

StudioCast does not commit model binaries into git.

Installed model packs live at:

${XDG_DATA_HOME:-$HOME/.local/share}/studiocast/models/open_video/

For sourcing, conversion, and hosting of the actual model binaries (ONNX / dlib data files), see:

docs/open_video_model_conversion.md

Subject areas

segmentation/ — matte/alpha models (task = "matting")
face_detection/ — face detection models
- Templates include YuNet in Good/Better/Best variants (int8 / int8bq / fp32).
face_landmarks/ — facial landmark detectors
video_denoise/ — video denoising models
- Templates include FastDVDnet sigma presets in Good/Better/Best variants (sigma15 / sigma25 / sigma50).
eye_contact/ — gaze/eye contact models
- Templates include gaze-correction-cam in Good/Better/Best variants (FP16 / FP32 / v3 placeholder).

What’s in a pack?

A pack is a directory containing:

model.json — pack metadata (see docs/open_video_model_packs.md)
LICENSE.txt — upstream licensing summary
model artifacts (downloaded at install time), e.g.
- *.onnx
- shape_predictor_68_face_landmarks.dat

Model info

onnx region:us

↓ 0 ♥ 0

Created: 2026-02-25

Updated: 2026-02-26

View on Hugging Face

Files (55)

.gitattributes

README.md

eye_contact/Best Quality (Gaze Correction Cam v3 Placeholder)/LICENSE.txt

eye_contact/Best Quality (Gaze Correction Cam v3 Placeholder)/model.json

eye_contact/Better Quality (Gaze Correction Cam v0.1.1 FP32)/LICENSE.txt

eye_contact/Better Quality (Gaze Correction Cam v0.1.1 FP32)/gaze_flx_left.onnx ONNX

eye_contact/Better Quality (Gaze Correction Cam v0.1.1 FP32)/gaze_flx_right.onnx ONNX

eye_contact/Better Quality (Gaze Correction Cam v0.1.1 FP32)/model.json

eye_contact/Good Quality (Gaze Correction Cam v0.1.1 FP16)/LICENSE.txt

eye_contact/Good Quality (Gaze Correction Cam v0.1.1 FP16)/gaze_flx_left_fp16.onnx ONNX

eye_contact/Good Quality (Gaze Correction Cam v0.1.1 FP16)/gaze_flx_right_fp16.onnx ONNX

eye_contact/Good Quality (Gaze Correction Cam v0.1.1 FP16)/model.json

face_detection/Best Quality (YuNet FP32)/LICENSE.txt

face_detection/Best Quality (YuNet FP32)/model.json

face_detection/Best Quality (YuNet FP32)/model.onnx ONNX

face_detection/Better Quality (YuNet INT8 BQ)/LICENSE.txt

face_detection/Better Quality (YuNet INT8 BQ)/model.json

face_detection/Better Quality (YuNet INT8 BQ)/model.onnx ONNX

face_detection/Good Quality (YuNet INT8)/LICENSE.txt

face_detection/Good Quality (YuNet INT8)/model.json

face_detection/Good Quality (YuNet INT8)/model.onnx ONNX

face_landmarks/Dlib 68 (iBUG 300-W)/LICENSE.txt

face_landmarks/Dlib 68 (iBUG 300-W)/model.json

face_landmarks/Dlib 68 (iBUG 300-W)/shape_predictor_68_face_landmarks.dat

matting/Best Quality/LICENSE.txt

matting/Best Quality/model.json

matting/Best Quality/model.onnx ONNX

matting/Better Quality/LICENSE.txt

matting/Better Quality/model.json

matting/Better Quality/model.onnx ONNX

matting/Good Quality/LICENSE.txt

matting/Good Quality/model.json

matting/Good Quality/model.onnx ONNX

open_audio/README.md

open_audio/fastenhancer_l_vd_v1/LICENSE.txt

open_audio/fastenhancer_l_vd_v1/README.txt

open_audio/fastenhancer_l_vd_v1/model.json

open_audio/fastenhancer_l_vd_v1/model.onnx ONNX

open_audio/fastenhancer_m_vd_v1/LICENSE.txt

open_audio/fastenhancer_m_vd_v1/README.txt

open_audio/fastenhancer_m_vd_v1/model.json

open_audio/fastenhancer_m_vd_v1/model.onnx ONNX

open_audio/fastenhancer_s_vd_v1/LICENSE.txt

open_audio/fastenhancer_s_vd_v1/README.txt

open_audio/fastenhancer_s_vd_v1/model.json

open_audio/fastenhancer_s_vd_v1/model.onnx ONNX

video_denoise/Best Quality (FastDVDnet sigma50)/LICENSE.txt

video_denoise/Best Quality (FastDVDnet sigma50)/model.json

video_denoise/Best Quality (FastDVDnet sigma50)/model.onnx ONNX

video_denoise/Better Quality (FastDVDnet sigma25)/LICENSE.txt

video_denoise/Better Quality (FastDVDnet sigma25)/model.json

video_denoise/Better Quality (FastDVDnet sigma25)/model.onnx ONNX

video_denoise/Good Quality (FastDVDnet sigma15)/LICENSE.txt

video_denoise/Good Quality (FastDVDnet sigma15)/model.json

video_denoise/Good Quality (FastDVDnet sigma15)/model.onnx ONNX

Related models

More from 10dallasj

No other models by this author.

More models

No task type declared.