Files
BGfilter/configs/default.yaml
T
lhk229 08378e57f8 Add anime-seg segmentation backend (SkyTNT ISNet)
Add AnimeSegSegmenter (skytnt/anime-seg ISNet ONNX via onnxruntime, no remote
code) and a make_segmenter factory selected by SegmentationSettings.backend
("birefnet" | "anime-seg"). The ONNX output is already 0..1, so it slots into the
same soft-mask interface BiRefNetSegmenter uses.

On the anime samples anime-seg recovers more and more-coherent hair wisps than
BiRefNet (TestImage2 shoulder rescue: added px 1886 -> 3278, largest connected
component 147 -> 454), as expected from an anime-trained model. Default backend
stays birefnet.

Adds onnxruntime to requirements.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-30 23:58:22 +08:00

68 lines
1.5 KiB
YAML

model:
model_name: hustvl/vitmatte-base-composition-1k
device: cuda
matting_method: vitmatte
fallback_to_chroma_alpha: false
segmentation:
enabled: true
backend: birefnet # birefnet | anime-seg (anime-seg needs model_name: skytnt/anime-seg)
model_name: ZhengPeng7/BiRefNet
device: cuda
input_size: 1024
trimap:
sure_bg_threshold: 0.92
sure_fg_threshold: 0.12
unknown_radius_ratio: 0.012
fg_safe_radius_ratio: 0.006
seg_core_threshold: 0.60
seg_loose_threshold: 0.08
alpha_post:
chroma_suppress: true
chroma_suppress_bg_low: 0.35
chroma_suppress_bg_high: 0.80
chroma_suppress_strength: 1.0
seg_fill: true
seg_fill_alpha_max: 0.10
seg_fill_seg_min: 0.10
seg_fill_strength: 0.70
seg_fill_min_area_ratio: 0.00002
foreground:
enabled: true
method: ml
fallback_to_unmix: true
ml_regularization: 0.00001
# Legacy "unmix" method params (used only when method: unmix).
edge_low: 0.005
edge_high: 0.995
min_unmix_alpha: 0.08
unmix_strength: 0.70
local_strength: 0.45
local_blur_radius: 11
local_alpha_threshold: 0.92
local_bg_confidence_max: 0.20
green_excess_margin: 0.015
bg_confidence_weight: 0.70
despill:
enabled: true
edge_low: 0.005
strength: 0.92
green_excess_margin: 0.015
edge_expand_radius: 2
alpha_weight_floor: 0.35
recolor:
enabled: true
person_threshold: 0.5
coverage_soft: 0.08
contam_low: 0.02
contam_high: 0.12
clean_dom_max: 0.03
propagate_radius_ratio: 0.02
min_reference_weight: 0.05
luminance_strength: 0.5