SnapOtterandGitHub 1bac663a2e feat(erase-object): optional high-quality diffusion inpainting bundle (#566) ...

Adds an opt-in High Quality mode to the Object Eraser, backed by a new inpaint-hq feature bundle (Stable Diffusion 1.5 inpainting via diffusers). The default fast LaMa path is unchanged. Both arch archives are published to deepsafe/feature-bundles and the manifest carries their real sha256/sizes.

Verified end to end: a fresh container pulls the bundle from HuggingFace, checksum-verifies it, extracts torch/diffusers plus the fp16 model, and the HQ sidecar erases a large object with a plausible fill.

Refs #141

2026-07-19 20:47:35 +08:00

..

models

feat(noise-removal): SOTA noise removal with 4 quality tiers (#57)

2026-04-13 19:50:23 +08:00

tests

feat(erase-object): optional high-quality diffusion inpainting bundle (#566)

2026-07-19 20:47:35 +08:00

colorize.py

fix(ai): surface actionable fix for libGL.so.1 missing on headless installs

2026-06-10 18:59:30 +08:00

detect_faces.py

fix: remove automatic third-party egress of user data + optional strict offline mode (OSM tiles, Scalar fonts, editor fonts, AI model downloads) (#422)

2026-07-04 05:46:52 +00:00

dispatcher.py

feat(erase-object): optional high-quality diffusion inpainting bundle (#566)

2026-07-19 20:47:35 +08:00

doc_flatten.py

feat(tools): 2.0 phase 5 wave 2 - pdf depth (21 tools) (#220)

2026-06-13 10:18:55 +08:00

doc_health.py

feat(modality)!: SnapOtter 2.0 phase 3 modality framework: media/doc engines, pool routing, display modes (#218)

2026-06-13 10:18:39 +08:00

doc_html_pdf.py

fix(ai): broaden SSRF pre-scan regex to cover srcset, poster, formaction, @import

2026-06-13 10:36:12 +08:00

doc_metadata.py

feat(tools): 2.0 phase 5 wave 2 - pdf depth (21 tools) (#220)

2026-06-13 10:18:55 +08:00

doc_pagecount.py

feat(modality)!: SnapOtter 2.0 phase 3 modality framework: media/doc engines, pool routing, display modes (#218)

2026-06-13 10:18:39 +08:00

doc_redact.py

fix: PDF tool QA sweep - library auto-save versioning, AI fileId threading, modality polish (#251)

2026-06-16 15:48:07 +08:00

doc_scrub_meta.py

fix: stamp SnapOtter as Producer on generated PDFs (#416)

2026-07-04 03:05:32 +00:00

doc_sign.py

feat: add Sign PDF tool (draw/type/upload signatures, place on a PDF) (#370)

2026-06-29 11:01:25 +08:00

doc_text.py

feat(tools): 2.0 phase 5 wave 2 - pdf depth (21 tools) (#220)

2026-06-13 10:18:55 +08:00

doc_to_word.py

fix: preserve colored blocks in PDF-to-Word (#500)

2026-07-11 19:20:32 +08:00

enhance_faces.py

fix(ai): gate AI tools on per-framework GPU detection, not a shared boolean (#445)

2026-07-06 18:39:01 +08:00

face_landmarks.py

fix: remove automatic third-party egress of user data + optional strict offline mode (OSM tiles, Scalar fonts, editor fonts, AI model downloads) (#422)

2026-07-04 05:46:52 +00:00

gif_remove_bg.py

feat(tools): remove background from animated GIFs (WebP, APNG) (#502)

2026-07-11 19:53:00 +08:00

gpu.py

fix(ai-bundles): stop CPU onnxruntime from clobbering onnxruntime-gpu (#544)

2026-07-17 00:25:32 +08:00

inpaint_hq.py

feat(erase-object): optional high-quality diffusion inpainting bundle (#566)

2026-07-19 20:47:35 +08:00

inpaint.py

fix(erase-object): crop-based HD inpainting to remove ghosting and blur (#501)

2026-07-11 22:27:35 +08:00

install_feature.py

fix(ai-bundles): stop CPU onnxruntime from clobbering onnxruntime-gpu (#544)

2026-07-17 00:25:32 +08:00

install_runtime.py

fix: make OCR portable and reliable across AMD64 and ARM64 (#519)

2026-07-15 03:34:24 +08:00

noise_removal.py

fix(ai): gate AI tools on per-framework GPU detection, not a shared boolean (#445)

2026-07-06 18:39:01 +08:00

ocr_preprocess.py

feat(ocr): add OpenCV preprocessing pipeline for OCR

2026-04-12 18:30:00 +08:00

ocr_runtime_entrypoint.py

fix: make OCR portable and reliable across AMD64 and ARM64 (#519)

2026-07-15 03:34:24 +08:00

ocr_runtime.py

fix: make OCR portable and reliable across AMD64 and ARM64 (#519)

2026-07-15 03:34:24 +08:00

offline_guard.py

fix: offline CodeFormer face-enhance (ship RealESRGAN_x2plus in upscale-enhance bundle) (#433)

2026-07-04 16:30:08 +00:00

outpaint.py

fix: remove automatic third-party egress of user data + optional strict offline mode (OSM tiles, Scalar fonts, editor fonts, AI model downloads) (#422)

2026-07-04 05:46:52 +00:00

pdf2docx_layout.py

fix: preserve colored blocks in PDF-to-Word (#500)

2026-07-11 19:20:32 +08:00

red_eye_removal.py

fix: remove automatic third-party egress of user data + optional strict offline mode (OSM tiles, Scalar fonts, editor fonts, AI model downloads) (#422)

2026-07-04 05:46:52 +00:00

remove_bg.py

fix: remove automatic third-party egress of user data + optional strict offline mode (OSM tiles, Scalar fonts, editor fonts, AI model downloads) (#422)

2026-07-04 05:46:52 +00:00

requirements-gpu.txt

fix: make OCR portable and reliable across AMD64 and ARM64 (#519)

2026-07-15 03:34:24 +08:00

requirements.txt

fix: make OCR portable and reliable across AMD64 and ARM64 (#519)

2026-07-15 03:34:24 +08:00

restore.py

fix(ai): gate AI tools on per-framework GPU detection, not a shared boolean (#445)

2026-07-06 18:39:01 +08:00

transcribe.py

fix(ai): gate AI tools on per-framework GPU detection, not a shared boolean (#445)

2026-07-06 18:39:01 +08:00

upscale.py

fix(ai): gate AI tools on per-framework GPU detection, not a shared boolean (#445)

2026-07-06 18:39:01 +08:00