• Implemented global offline mode for all HuggingFace-dependent modules to eliminate unauthenticated request warnings and prevent external connection attempts during startup and runtime.
Changes
• Global Offline Configuration: Added HF_HUB_OFFLINE=1 and TRANSFORMERS_OFFLINE=1 to the environment variables at the top of edit_api.py, embeddings.py, watcher.py, and pose_llm/pose_llm_api.py.
• Explicit Local Files Only: Updated all huggingface_hub.hf_hub_download calls in edit_api.py to include local_files_only=True, ensuring they never attempt to reach the Hub if models are already cached.
• LLM Service Optimization: Updated AutoTokenizer and AutoModelForCausalLM calls in pose_llm/pose_llm_api.py to use local_files_only=True.
• Consistency: Ensured that shared modules like embeddings.py which uses open_clip are also locked to offline mode to prevent background downloads.
• Implemented global offline mode for all HuggingFace-dependent modules to eliminate unauthenticated request warnings and prevent external connection attempts during startup and runtime.
Changes
• Global Offline Configuration: Added HF_HUB_OFFLINE=1 and TRANSFORMERS_OFFLINE=1 to the environment variables at the top of edit_api.py, embeddings.py, watcher.py, and pose_llm/pose_llm_api.py.
• Explicit Local Files Only: Updated all huggingface_hub.hf_hub_download calls in edit_api.py to include local_files_only=True, ensuring they never attempt to reach the Hub if models are already cached.
• LLM Service Optimization: Updated AutoTokenizer and AutoModelForCausalLM calls in pose_llm/pose_llm_api.py to use local_files_only=True.
• Consistency: Ensured that shared modules like embeddings.py which uses open_clip are also locked to offline mode to prevent background downloads.
• Implemented global offline mode for all HuggingFace-dependent modules to eliminate unauthenticated request warnings and prevent external connection attempts during startup and runtime.
Changes
• Global Offline Configuration: Added HF_HUB_OFFLINE=1 and TRANSFORMERS_OFFLINE=1 to the environment variables at the top of edit_api.py, embeddings.py, watcher.py, and pose_llm/pose_llm_api.py.
• Explicit Local Files Only: Updated all huggingface_hub.hf_hub_download calls in edit_api.py to include local_files_only=True, ensuring they never attempt to reach the Hub if models are already cached.
• LLM Service Optimization: Updated AutoTokenizer and AutoModelForCausalLM calls in pose_llm/pose_llm_api.py to use local_files_only=True.
• Consistency: Ensured that shared modules like embeddings.py which uses open_clip are also locked to offline mode to prevent background downloads.
• Implemented global offline mode for all HuggingFace-dependent modules to eliminate unauthenticated request warnings and prevent external connection attempts during startup and runtime.
Changes
• Global Offline Configuration: Added HF_HUB_OFFLINE=1 and TRANSFORMERS_OFFLINE=1 to the environment variables at the top of edit_api.py, embeddings.py, watcher.py, and pose_llm/pose_llm_api.py.
• Explicit Local Files Only: Updated all huggingface_hub.hf_hub_download calls in edit_api.py to include local_files_only=True, ensuring they never attempt to reach the Hub if models are already cached.
• LLM Service Optimization: Updated AutoTokenizer and AutoModelForCausalLM calls in pose_llm/pose_llm_api.py to use local_files_only=True.
• Consistency: Ensured that shared modules like embeddings.py which uses open_clip are also locked to offline mode to prevent background downloads.
• Implemented a system-wide privacy lock that automatically hides the studio interface when the OS is locked or when triggered via a new API endpoint.
Changes
• Backend edit_api.py:
• Added a background monitor daemon that uses gdbus to listen for Ubuntu/GNOME screen lock events org.gnome.ScreenSaver.ActiveChanged.
• Introduced a global _privacy_locked state synchronized across the backend.
• Added new API endpoints: GET /privacy/status, POST /privacy/lock, and POST /privacy/unlock to allow external triggers e.g., keyboard macros.
• Updated the static data exporter to include system_status.json, enabling efficient frontend polling.
• Frontend car.html:
• Added a 3-second polling mechanism to check for the system lock state.
• Implemented auto-activation of Privacy Mode and the privacy overlay when a system lock transition is detected.
• Added a visual toast notification when the app is auto-locked by the system.
Verification
• Verified backend code integrity via py_compile.
• Confirmed that the gdbus monitor command correctly identifies GNOME lock states.
• Ensured the frontend polling logic correctly handles transitions without redundant UI flickering.
Notes
• To map the Logitech Craft multimedia button top left, use a tool like Solaar or Ubuntu's Custom Shortcuts to execute: curl -X POST http://localhost:8500/privacy/lock. This will instantly hide the app regardless of browser focus.
• Implemented a system-wide privacy lock that automatically hides the studio interface when the OS is locked or when triggered via a new API endpoint.
Changes
• Backend edit_api.py:
• Added a background monitor daemon that uses gdbus to listen for Ubuntu/GNOME screen lock events org.gnome.ScreenSaver.ActiveChanged.
• Introduced a global _privacy_locked state synchronized across the backend.
• Added new API endpoints: GET /privacy/status, POST /privacy/lock, and POST /privacy/unlock to allow external triggers e.g., keyboard macros.
• Updated the static data exporter to include system_status.json, enabling efficient frontend polling.
• Frontend car.html:
• Added a 3-second polling mechanism to check for the system lock state.
• Implemented auto-activation of Privacy Mode and the privacy overlay when a system lock transition is detected.
• Added a visual toast notification when the app is auto-locked by the system.
Verification
• Verified backend code integrity via py_compile.
• Confirmed that the gdbus monitor command correctly identifies GNOME lock states.
• Ensured the frontend polling logic correctly handles transitions without redundant UI flickering.
Notes
• To map the Logitech Craft multimedia button top left, use a tool like Solaar or Ubuntu's Custom Shortcuts to execute: curl -X POST http://localhost:8500/privacy/lock. This will instantly hide the app regardless of browser focus.
New _apply_transparency_black_bg function (called when bg_removal=sam2):
1.
Threshold — any pixel with max-channel > 25 = person. Finds the person's exact bounding box without any model confusion.
2.
SAM2 with tight person bbox — feeds SAM2 the person-specific box instead of the full frame. SAM2 now segments within the person area for clean sub-pixel edges.
3.
Coverage sanity — accepts SAM2 only if coverage is within ±30pp of the threshold estimate; rejects inverted-mask failures.
4.
Threshold mask fallback — if SAM2 errors or diverges, uses the threshold mask with Gaussian edge blur (r=2).
Test result: Person RGB mean (146, 101, 86) — correct skin tones. 74.5% transparent background, 24% opaque person. ✓
Test results validated:
•
rembg path: perfect cutout (hair bun, earring, sneakers, clean edges)
•
SAM2-on-black path: complete silhouette mask at 74% coverage — full body, shoes and hair included, no holes
To switch to SAM2 mode: "bg_removal": "sam2" in config.json. No restart needed — the config is read per-request.
New _apply_transparency_black_bg function (called when bg_removal=sam2):
1.
Threshold — any pixel with max-channel > 25 = person. Finds the person's exact bounding box without any model confusion.
2.
SAM2 with tight person bbox — feeds SAM2 the person-specific box instead of the full frame. SAM2 now segments within the person area for clean sub-pixel edges.
3.
Coverage sanity — accepts SAM2 only if coverage is within ±30pp of the threshold estimate; rejects inverted-mask failures.
4.
Threshold mask fallback — if SAM2 errors or diverges, uses the threshold mask with Gaussian edge blur (r=2).
Test result: Person RGB mean (146, 101, 86) — correct skin tones. 74.5% transparent background, 24% opaque person. ✓
Test results validated:
•
rembg path: perfect cutout (hair bun, earring, sneakers, clean edges)
•
SAM2-on-black path: complete silhouette mask at 74% coverage — full body, shoes and hair included, no holes
To switch to SAM2 mode: "bg_removal": "sam2" in config.json. No restart needed — the config is read per-request.
