* Added degirum plugin, updated documentation for degirum detector usage, updated requirements with degirum_headless
* Fixed broken link
* Made it so openvino prioritizes using GPU and NPU over CPU
* Version that detects model and can begin using @local
* Updating requirements to build dev container
* Added optimized version of degirum plugin + updated docs
* Added guard clause for empty inference reponse
* Updated DeGirum's docs
* Moved DeGirum section to 'Community' detectors, fixed formatting of headers to be more consistent with the rest of the page, and removed uneeded 'models' folder
* Moved DeGirum section to correct place in community models
* Update ROCm to 6.4.0 (#18264)
* Update to rocm 6.4.0
* Update URL
* Remove old env var
* Dynamic Config Updates (#18353)
* Create classes to handle publishing and subscribing config updates
* Cleanup
* Use config updater
* Update handling for enabled config
* Cleanup
* Recording config updates
* Birdseye config updates
* Handle notifications
* handle review
* Update motion
* Dynamically update masks and zones for cameras (#18359)
* Include config publisher in api
* Call update topic for passed topics
* Update zones dynamically
* Update zones internally
* Support zone and mask reset
* Handle updating objects config
* Don't put status for needing to restart Frigate
* Cleanup http tests
* Fix tests
* Initial custom classification model config support (#18362)
* Add basic config for defining a teachable machine model
* Add model type
* Add basic config for teachable machine models
* Adjust config for state and object
* Use config to process
* Correctly check for objects
* Remove debug
* Rename to not be teachable machine specific
* Cleanup
* Implement support for no recordings indicator on timeline (#18363)
* Indicate no recordings on the history timeline with gray hash marks
This commit includes a new backend API endpoint and the frontend changes needed to support this functionality
* don't show slashes for now
* Update ROCm to 6.4.1 (#18364)
* Update rocm to 6.4.1
* Quick fix
* Add ability to configure when custom classification models run (#18380)
* Add config to control when classification models are run
* Cleanup
* Add basic config editor when Frigate can't startup (#18383)
* Start Frigate in safe mode when config does not validate
* Add safe mode page that is just the config editor
* Adjust Frigate config editor when in safe mode
* Cleanup
* Improve log message
* Fix incorrectly running lpr (#18390)
* Audio transcription support (#18398)
* install new packages for transcription support
* add config options
* audio maintainer modifications to support transcription
* pass main config to audio process
* embeddings support
* api and transcription post processor
* embeddings maintainer support for post processor
* live audio transcription with sherpa and faster-whisper
* update dispatcher with live transcription topic
* frontend websocket
* frontend live transcription
* frontend changes for speech events
* i18n changes
* docs
* mqtt docs
* fix linter
* use float16 and small model on gpu for real-time
* fix return value and use requestor to embed description instead of passing embeddings
* run real-time transcription in its own thread
* tweaks
* publish live transcriptions on their own topic instead of tracked_object_update
* config validator and docs
* clarify docs
* Implement API to train classification models (#18475)
* Intel updates (#18493)
* Update openvino and onnxruntime
* Install icd and level-zero-gpu deps from intel directly
* Install
* Add dep
* Fix package install
* Tiered recordings (#18492)
* Implement tiered recording
* Add migration for record config
* Update docs
* Update reference docs
* Fix preview query
* Fix incorrect accesses
* Fix
* Fix
* Fix
* Fix
* Upgrade PaddleOCR models to v4 (rec) and v5 (det) (#18505)
The PP_OCRv5 text detection models have greatly improved over v3. The v5 recognition model makes improvements to challenging handwriting and uncommon characters, which are not necessary for LPR, so using v4 seemed like a better choice to continue to keep inference time as low as possible. Also included is the full dictionary for Chinese character support.
* Audio transcription tweaks (#18540)
* use model runner
* unload whisper model when live transcription is complete
* Classification Model UI (#18571)
* Setup basic training structure
* Build out route
* Handle model configs
* Add image fetch APIs
* Implement model training screen with dataset selection
* Implement viewing of training images
* Adjust directories
* Implement viewing of images
* Add support for deleting images
* Implement full deletion
* Implement classification model training
* Improve naming
* More renaming
* Improve layout
* Reduce logging
* Cleanup
* Live classification model training (#18583)
* Implement model training via ZMQ and add model states to represent training
* Get model updates working
* Improve toasts and model state
* Clean up logging
* Add back in
* Classification Model Metrics (#18595)
* Add speed and rate metrics for custom classification models
* Use metrics for classification models
* Use keys
* Cast to list
* Add Mesa Teflon as a TFLite detector (#18310)
* Refactor common functions for tflite detector implementations
* Add detector using mesa teflon delegate
Non-EdgeTPU TFLite can use the standard .tflite format
* Add mesa-teflon-delegate from bookworm-backports to arm64 images
* feat: enable using GenAI for cameras with GenAI disabled from the API (#18616)
* fix: Initialize GenAI client if GenAI is enabled globally (#18623)
* Make Birdseye clickable (#18628)
* keep track of layout changes and publish on change
* websocket hook
* clickable overlay div to navigate to full camera view
* Refactor TensorRT (#18643)
* Combine base and arm trt detectors
* Remove unused deps for amd64 build
* Add missing packages and cleanup ldconfig
* Expand packages for tensorflow model training
* Cleanup
* Refactor training to not reserve memory
* Dynamic Management of Cameras (#18671)
* Add base class for global config updates
* Add or remove camera states
* Move camera process management to separate thread
* Move camera management fully to separate class
* Cleanup
* Stop camera processes when stop command is sent
* Start processes dynamically when needed
* Adjust
* Leave extra room in tracked object queue for two cameras
* Dynamically set extra config pieces
* Add some TODOs
* Fix type check
* Simplify config updates
* Improve typing
* Correctly handle indexed entries
* Cleanup
* Create out SHM
* Use ZMQ for signaling object detectoin is completed
* Get camera correctly created
* Cleanup for updating the cameras config
* Cleanup
* Don't enable audio if no cameras have audio transcription
* Use exact string so similar camera names don't interfere
* Add ability to update config via json body to config/set endpoint
Additionally, update the config in a single rather than multiple calls for each updated key
* fix autotracking calibration to support new config updater function
---------
Co-authored-by: Josh Hawkins <32435876+hawkeye217@users.noreply.github.com>
* Use Fork-Server As Spawn Method (#18682)
* Set runtime
* Use count correctly
* Don't assume camera sizes
* Use separate zmq proxy for object detection
* Correct order
* Use forkserver
* Only store PID instead of entire process reference
* Cleanup
* Catch correct errors
* Fix typing
* Remove before_run from process util
The before_run never actually ran because:
You're right to suspect an issue with before_run not being called and a potential deadlock. The way you've implemented the run_wrapper using __getattribute__ for the run method of BaseProcess is a common pitfall in Python's multiprocessing, especially when combined with how multiprocessing.Process works internally.
Here's a breakdown of why before_run isn't being called and why you might be experiencing a deadlock:
The Problem: __getattribute__ and Process Serialization
When you create a multiprocessing.Process object and call start(), the multiprocessing module needs to serialize the process object (or at least enough of it to re-create the process in the new interpreter). It then pickles this serialized object and sends it to the newly spawned process.
The issue with your __getattribute__ implementation for run is that:
run is retrieved during serialization: When multiprocessing tries to pickle your Process object to send to the new process, it will likely access the run attribute. This triggers your __getattribute__ wrapper, which then tries to bind run_wrapper to self.
run_wrapper is bound to the parent process's self: The run_wrapper closure, when created in the parent process, captures the self (the Process instance) from the parent's memory space.
Deserialization creates a new object: In the child process, a new Process object is created by deserializing the pickled data. However, the run_wrapper method that was pickled still holds a reference to the self from the parent process. This is a subtle but critical distinction.
The child's run is not your wrapped run: When the child process starts, it internally calls its own run method. Because of the serialization and deserialization process, the run method that's ultimately executed in the child process is the original multiprocessing.Process.run or the Process.run if you had directly overridden it. Your __getattribute__ magic, which wraps run, isn't correctly applied to the Process object within the child's context.
* Cleanup
* Logging bugfix (#18465)
* use mp Manager to handle logging queues
A Python bug (https://github.com/python/cpython/issues/91555) was preventing logs from the embeddings maintainer process from printing. The bug is fixed in Python 3.14, but a viable workaround is to use the multiprocessing Manager, which better manages mp queues and causes the logging to work correctly.
* consolidate
* fix typing
* Fix typing
* Use global log queue
* Move to using process for logging
* Convert camera tracking to process
* Add more processes
* Finalize process
* Cleanup
* Cleanup typing
* Formatting
* Remove daemon
---------
Co-authored-by: Josh Hawkins <32435876+hawkeye217@users.noreply.github.com>
* Add basic camera settings to UI for testing (#18690)
* add basic camera add/edit pane to the UI for testing
* only init model runner if transcription is enabled globally
* fix role checkboxes
* Ensure logging config is propagated to forked processes (#18704)
* Move log level initialization to log
* Use logger config
* Formatting
* Fix config order
* Set process names
---------
Co-authored-by: Nicolas Mowen <nickmowen213@gmail.com>
* Fix go2rtc init (#18708)
* Cleanup process handling
* Adjust process name
* Reduce tf initialization
* Don't use staticmethod
* Don't fail on unicode debug for config updates
* Catch unpickling error
* Fix birdseye crash when dynamically adding a camera (#18821)
* Catch invalid character index in lpr CTC decoder (#18825)
* Classification model cover images (#18843)
* Move to separate component
* Add cover images for clssification models
* Fix process name
* Handle SIGINT with forkserver (#18860)
* Pass stopevent from main start
* Share stop event across processes
* preload modules
* remove explicit os._exit call
---------
Co-authored-by: Josh Hawkins <32435876+hawkeye217@users.noreply.github.com>
* Don't try to close or join mp manager queues (#18866)
Multiprocessing Manager queues don't have a close() or join_thread() method, and the Manager will clean it up appropriately after we empty it. This prevents an infinite loop when an AttributeError exception fires for Manager AutoProxy queue objects.
* Improve logging (#18867)
* Ignore numpy get limits warning
* Add function wrapper to redirect stdout and stderr to logpipe
* Save stderr too
* Add more to catch
* run logpipe
* Use other logging redirect class
* Use other logging redirect class
* add decorator for redirecting c/c++ level output to logger
* fix typing
---------
Co-authored-by: Josh Hawkins <32435876+hawkeye217@users.noreply.github.com>
* Add ONVIF focus support (#18883)
* backend
* frontend and i18n
* 0.17 tweaks (#18892)
* Set version
* Cleanup more logs
* Don't log matplotlib
* Improve object classification (#18908)
* Ui improvements
* Improve image cropping and model saving
* Improve naming
* Add logs for training
* Improve model labeling
* Don't set sub label for none object classification
* Cleanup
* Remove TFLite init logs
* Improve classification UI (#18910)
* Move threhsold to base model config
* Improve score handling
* Add back button
* Classification improvements (#19020)
* Move classification training to full process
* Sort class images
* Semantic Search Triggers (#18969)
* semantic trigger test
* database and model
* config
* embeddings maintainer and trigger post-processor
* api to create, edit, delete triggers
* frontend and i18n keys
* use thumbnail and description for trigger types
* image picker tweaks
* initial sync
* thumbnail file management
* clean up logs and use saved thumbnail on frontend
* publish mqtt messages
* webpush changes to enable trigger notifications
* add enabled switch
* add triggers from explore
* renaming and deletion fixes
* fix typing
* UI updates and add last triggering event time and link
* log exception instead of return in endpoint
* highlight entry in UI when triggered
* save and delete thumbnails directly
* remove alert action for now and add descriptions
* tweaks
* clean up
* fix types
* docs
* docs tweaks
* docs
* reuse enum
* Optionally show tracked object paths in debug view (#19025)
* Dynamically enable/disable GenAI (#19139)
* config
* dispatcher and mqtt
* docs
* use config updater
* add switch to frontend
* Classification train updates (#19173)
* Improve model train button
* Add filters for classification
* Cleanup
* Don't run classification on false positives
* Cleanup filter
* Fix icon color
* Object attribute classification (#19205)
* Add enum for type of classification for objects
* Update recognized license plate topic to be used as attribute updater
* Update attribute for attribute type object classification
* Cleanup
* Require setting process priority for FrigateProcess (#19207)
* Add bookworm-backports to the rocm images and upgrade mesa/vaapi to support RDNA4 GPUs (#19312)
* Improve the tablet layout (#19320)
* Improve the tablet layout
* Update imports sort
* Fix more imports
* Implement start for review item description processor (#19352)
* Add review item data transmission
* Publish review updates
* Add review item subscriber
* Basic implementation for testing review processor
* Formatting
* Cleanup
* Improve comms typing (#18599)
* Enable mypy for comms
* Make zmq data types consistent
* Cleanup inter process typing issues
* Cleanup embeddings typing
* Cleanup config updater
* Cleanup recordings updator
* Make publisher have a generic type
* Cleanup event metadata updater
* Cleanup event metadata updater
* Cleanup detections updater
* Cleanup websocket
* Cleanup mqtt
* Cleanup webpush
* Cleanup dispatcher
* Formatting
* Remove unused
* Add return type
* Fix tests
* Fix semantic triggers config typing
* Cleanup
* Ensure alertVideos persistence is loaded before displaying thumb or preview (#19432)
The default value of true would cause previews to be loaded in the background even if the local storage value was false
* Adjust loitering behavior based on object type (#19433)
* Adjust loitering behavior based on object
* Update docs
* Grammar
* Enable mypy for DB and fix types (#19434)
* Install peewee type hints
* Models now have proper types
* Fix iterator type
* Enable debug builds with dev reqs installed
* Install as wheel
* Fix cast type
* Migrate object genai configuration (#19437)
* Move genAI object to objects section
* Adjust config propogation behavior
* Refactor genai config usage
* Automatic migration
* Always start the embeddings process
* Always init embeddings
* Config fixes
* Adjust reference config
* Adjust docs
* Formatting
* Fix
* Review Item GenAI metadata (#19442)
* Rename existing function
* Keep track of thumbnial updates
* Tinkering with genai prompt
* Adjust input format
* Create model for review description output
* testing prompt changes
* Prompt improvements and image saving
* Add config for review items genai
* Use genai review config
* Actual config usage
* Adjust debug image saving
* Fix
* Fix review creation
* Adjust prompt
* Prompt adjustment
* Run genai in thread
* Fix detections block
* Adjust prompt
* Prompt changes
* Save genai response to metadata model
* Handle metadata
* Send review update to dispatcher
* Save review metadata to DB
* Send review notification updates
* Quick fix
* Fix name
* Fix update type
* Correctly dump model
* Add card
* Add card
* Remove message
* Cleanup typing and UI
* Adjust prompt
* Formatting
* Add log
* Formatting
* Add inference speed and keep alive
* Review genai updates (#19448)
* Include extra level for normal activity
* Add dynamic toggling
* Update docs
* Add different threshold for genai
* Adjust webUI for object and review description feature
* Adjust config
* Send on startup
* Cleanup config setting
* Set config
* Fix config name
* Use preview frames for Review Descriptions (#19450)
* Use preview frames for genai
* Cleanup
* Adjust
* Add config for users to define additional concerns that GenAI should make note of in review summary (#19463)
* Don't default to openai
* Improve UI
* Allow configuring additional concerns that users may want the AI to note
* Formatting
* Add preferred language config
* Remove unused
* Added total camera fps, total processed fps, and total skipped fps to stats api (#19469)
Co-authored-by: Mark Francis <markfrancisonly@gmail.com>
* Genai review summaries (#19473)
* Generate review item summaries with requests
* Adjust logic to only send important items
* Don't mention ladder
* Adjust prompt to be more specific
* Add more relaxed nature for normal activity
* Cleanup summary
* Update ollama client
* Add more directions to analyze the frames in order
* Remove environment from prompt
* Add ability to pass additional args to Ollama (#19484)
* Call out recognized objects more specifically
* Cleanup
* Make keep_alive and options configurable
* Generalize
* Use for other providers
* Update GenAI docs for new review summaries feature (#19493)
* Remove old genai docs
* Separate existing genai docs to separate sections
* Add docs for genai features
* Update reference config
* Update link
* Move to bottom
* Improve natural language of prompt (#19515)
* Make sequence details human-readable so they are used in natural language response
* Cleanup
* Improve prompt and image selection
* Adjust
* Adjust sligtly
* Format time
* Adjust frame selection logic
* Debug save response
* Ignore extra fields
* Adjust docs
* Cleanup filename sanitization
* Added degirum plugin, updated documentation for degirum detector usage, updated requirements with degirum_headless
* Fixed broken link
* Made it so openvino prioritizes using GPU and NPU over CPU
* Version that detects model and can begin using @local
* Added optimized version of degirum plugin + updated docs
* Updating requirements to build dev container
* Added guard clause for empty inference reponse
* Updated DeGirum's docs
* Moved DeGirum section to 'Community' detectors, fixed formatting of headers to be more consistent with the rest of the page, and removed uneeded 'models' folder
* Moved DeGirum section to correct place in community models
* Added degirum plugin, updated documentation for degirum detector usage, updated requirements with degirum_headless
* Fixed broken link
* Made it so openvino prioritizes using GPU and NPU over CPU
* Version that detects model and can begin using @local
* Added optimized version of degirum plugin + updated docs
* Updating requirements to build dev container
* Added guard clause for empty inference reponse
* Updated DeGirum's docs
* Moved DeGirum section to 'Community' detectors, fixed formatting of headers to be more consistent with the rest of the page, and removed uneeded 'models' folder
* Moved DeGirum section to correct place in community models
* Added degirum plugin, updated documentation for degirum detector usage, updated requirements with degirum_headless
* Fixed broken link
* Made it so openvino prioritizes using GPU and NPU over CPU
* Version that detects model and can begin using @local
* Added optimized version of degirum plugin + updated docs
* Updating requirements to build dev container
* Added guard clause for empty inference reponse
* Updated DeGirum's docs
* Moved DeGirum section to 'Community' detectors, fixed formatting of headers to be more consistent with the rest of the page, and removed uneeded 'models' folder
* Moved DeGirum section to correct place in community models
* Reverted changes to classification and audio
---------
Co-authored-by: Nicolas Mowen <nickmowen213@gmail.com>
Co-authored-by: Josh Hawkins <32435876+hawkeye217@users.noreply.github.com>
Co-authored-by: Jimmy <honj@alum.rpi.edu>
Co-authored-by: FL42 <46161216+fl42@users.noreply.github.com>
Co-authored-by: Steve Smith <tarkasteve@gmail.com>
Co-authored-by: markfrancisonly <12145270+markfrancisonly@users.noreply.github.com>
Co-authored-by: Mark Francis <markfrancisonly@gmail.com>
* Implement base rknn conversion
* Remove unused
* Formatting
* Add model conversion lock so it doesn't break when multiple detectors are defined
* Ignore unused impor
t
* Ignore numpy get limits warning
* Add function wrapper to redirect stdout and stderr to logpipe
* Save stderr too
* Add more to catch
* run logpipe
* Use other logging redirect class
* Use other logging redirect class
* add decorator for redirecting c/c++ level output to logger
* fix typing
---------
Co-authored-by: Josh Hawkins <32435876+hawkeye217@users.noreply.github.com>
* Refactor common functions for tflite detector implementations
* Add detector using mesa teflon delegate
Non-EdgeTPU TFLite can use the standard .tflite format
* Add mesa-teflon-delegate from bookworm-backports to arm64 images
* Don't support tensorrt detector for amd64 builds
* Add logs for directing users not to use tensorrt detector
* Rework docs
* Fix dockerfile index
* Don't undo jetson fix
Model can be adjusted ahead of time to NHWC to avoid transpose on CPU,
for example. All the model information is already presented in the
configuration, and the stringent checks implemented in openvino are not
present on other backends anyway. OpenVINO will properly report issues
with mismatched layouts anyhow.
* Don't return weighted name if it has the same number of results
* Remove link to incorrect format yolov9 models
* Fix command list from appearing when other inputs are focused
the description box in the tracked object details pane was causing the command input list to show when focused.
* clarify face docs
* Add note about python yolov9 export
* Check if hailort thread is still alive when timeout error is run into
* Reduce inference timeout
---------
Co-authored-by: Josh Hawkins <32435876+hawkeye217@users.noreply.github.com>
* improve spacing of face selection in mobile drawer
* fix spacing
* sort face names alphabetically
* Improve face selection dialog
* Use a state to track when face image loads
The naturalWidth and naturalHeight will always be 0 until the image loads. So we use onLoad and a state to track loading and then calculate the area after it has loaded
* Verify that a camera only tracks objects that are possible to track
* Fix test
* genai docs tweak
* Disable openvino model cache
* Clenaup
* Sanitize floats for estimated speed and angle
Users can configure speed zones in such a way that velocity estimates from Norfair cause a value of inf to be stored as an estimated speed. FastAPI doesn't serialize inf as a float, so trying to return this value would result in an API error. Sanitizing the value before storing should correct this.
---------
Co-authored-by: Nicolas Mowen <nickmowen213@gmail.com>
* Fix the `Any` typing hint treewide
There has been confusion between the Any type[1] and the any function[2]
in typing hints.
[1] https://docs.python.org/3/library/typing.html#typing.Any
[2] https://docs.python.org/3/library/functions.html#any
* Fix typing for various frame_shape members
Frame shapes are most likely defined by height and width, so a single int
cannot express that.
* Wrap gpu stats functions in Optional[]
These can return `None`, so they need to be `Type | None`, which is what
`Optional` expresses very nicely.
* Fix return type in get_latest_segment_datetime
Returns a datetime object, not an integer.
* Make the return type of FrameManager.write optional
This is necessary since the SharedMemoryFrameManager.write function can
return None.
* Fix total_seconds() return type in get_tz_modifiers
The function returns a float, not an int.
https://docs.python.org/3/library/datetime.html#datetime.timedelta.total_seconds
* Account for floating point results in to_relative_box
Because the function uses division the return types may either be int or
float.
* Resolve ruff deprecation warning
The config has been split into formatter and linter, and the global
options are deprecated.
* Add camera name tooltip to previews in recording view
* Apply face area check to cv2 face detection
* Delete review thumbnails
* Don't import hailo until it is used
* Add comment
* Clean up camera name
* Filter out empty keys when updating yaml config
HA ingress seems to randomly add an equal sign to the PUT urls for updating the config from the UI. This fix prevents empty keys from being processed, but still allows empty values.
---------
Co-authored-by: Nicolas Mowen <nickmowen213@gmail.com>
* face library i18n fixes
* face library i18n fixes
* add ability to use ctrl/cmd S to save in the config editor
* Use datetime as ID
* Update metrics inference speed to start with 0 ms
* fix android formatted thumbnail
* ensure role is comma separated and stripped correctly
* improve face library deletion
- add a confirmation dialog
- add ability to select all / delete faces in collections
* Implement lazy loading for video previews
* Force GPU for large embedding model
* GPU is required
* settings i18n fixes
* Don't delete train tab
* webpush debugging logs
* Fix incorrectly copying zones
* copy path data
* Ensure that cache dir exists for Frigate+
* face docs update
* Add description to upload image step to clarify the image
* Clean up
---------
Co-authored-by: Nicolas Mowen <nickmowen213@gmail.com>
* Adding Models
* Final Async Update
* Bug Fixing
* Fix
* Adding fixes
* Working async infer
* Final Documenatation and debug update
* Removing some extra prints
* Post-process correct label push
* config docs fix
* Review Fix
* Review fix 2.0
* Fixing the ASYNC API to work from 30ms to 10ms
* Fix for multi stream async infernce
* Format
* Fix#3
* Format#2
* Remove Unnessery includes
* Sort Imports
* Simplify rocm install and update to 6.3.1
* Build out more necessary packages
* Update to 6.3.3
* Set bake version
* Fix typo
* Ensure NHWC is used
* Reset dev changes
* Write to cache
* WIP yolov9
* Implement post processing for yolov9
* Cleanup detection
* Update docs to make note of supported yolov9
* Move post processing to separate utility
* Add note about other models
* rockchip: update dependencies and add script for model conversion
* rockchip: update docs
---------
Co-authored-by: Nicolas Mowen <nickmowen213@gmail.com>
