Commit Graph

52 Commits

Author SHA1 Message Date
Nicolas Mowen
18bf7f93fa
Improve rocm handling of different models (#14072)
* Improve rocm handling of different models

* Formatting

* Fix type check
2024-09-30 16:40:46 -05:00
gtsiam
c0bd3b362c
Custom classes for Process and Metrics (#13950)
* Subclass Process for audio_process

* Introduce custom mp.Process subclass

In preparation to switch the multiprocessing startup method away from
"fork", we cannot rely on os.fork cloning the log state at fork time.
Instead, we have to set up logging before we run the business logic of
each process.

* Make camera_metrics into a class

* Make ptz_metrics into a class

* Fixed PtzMotionEstimator.ptz_metrics type annotation

* Removed pointless variables

* Do not start audio processor when no audio cameras are configured
2024-09-27 07:53:23 -05:00
gtsiam
dc54981784
Move more things out of FrigateApp (#13897)
* Moved FrigateApp.init_config() into FrigateConfig.load()

* Move frigate config loading into main

* Store PlusApi in FrigateConfig

* Register SIGTERM handler in main

* Ensure logging is setup during config parsing

* Removed pointless try

* Moved config initialization out of FrigateApp

* Made FrigateApp.shm_frame_count into a function

* Removed log calls from signal handlers

python's logging calls are not re-entrant, which caused at least one of
these to deadlock randomly.

* Reopen stdout/err on process fork

This helps avoid deadlocks (https://github.com/python/cpython/issues/91776).

* Make mypy happy

* Whoops. I might have forgotten to save.

Truly an amateur mistake.

* Always call FrigateApp.stop()
2024-09-24 07:07:47 -05:00
Nicolas Mowen
06ccf7e9e9
Always close connection to shm frame after detection (#13766) 2024-09-16 07:56:20 -06:00
Nicolas Mowen
9afa1354da
Limit shm frame count (#12363)
* Limited shm frame count (#12346)

* Only keep 2x detect fps frames in SHM

* Don't delete previous shm frames in output

* Catch case where images do not exist

* Ensure files are closed

* Clear out all frames when shutting down

* Correct the number of frames saved

* Simplify empty shm error handling

* Improve frame safety

* Add handler logs when frame is None

* Don't fail on cleanup

* Cleanup logging

* Update docs

* Update calculation

* Restore condition

* Fix case where thumbnail is saved without frame

* Adjust debug logs

* Calculate best shm frame count

* Fix shm count calculation

* Catch missing frame

* Formatting

* Clarify docs

* Catch none frame in autotracking
2024-09-03 10:22:30 -06:00
Aida Syoko
3d612e510c
chore: fix some typos in comments (#11028)
Signed-off-by: alongdate <alongyear@outlook.com>
2024-04-20 06:16:43 -05:00
Nicolas Mowen
662025a961
Remove faster_fifo (#7181)
* Remove faster_fifo

* Remove const
2023-07-16 07:42:56 -05:00
Nate Meyer
5bb5e2dc5a
Add check for valid label index for object detection (#7149) (#7164) 2023-07-14 19:01:53 -05:00
Nicolas Mowen
baf671b764
Cleanup and organize utils (#7033)
* Force birdseye cameras into standard aspect ratios

* Organize utils

* Update tests

* Formatting

* Isort

* Fix tests

* Cleanup

* isort
2023-07-06 09:28:50 -05:00
Sergey Krashevich
c38c981cd0
Performance: multiprocessing improvement: step 2 (#6986)
* Refactored queues to use faster_fifo instead of mp.Queue

* Refactored LimitedQueue to include a counter for the number of items in the queue and updated put and get methods to use the counter

* Refactor app.py and util.py to use a custom Queue implementation called LQueue instead of the existing Queue

* Refactor put and get methods in LimitedQueue to handle queue size and blocking behavior more efficiently

* code format

* remove code from other branch (merging fuckup)
2023-07-06 07:56:38 -05:00
Blake Blackshear
ca7853c087
Fix max_frames, improve stationary objects in masked areas (#6815)
* fix issue with max_frames

* dont consider stationary until the threshold

* require a stationary interval

* try to fix formatter issues
2023-06-16 07:32:43 -05:00
Martin Weinelt
ab50d0b006
Add isort and ruff linter (#6575)
* Add isort and ruff linter

Both linters are pretty common among modern python code bases.

The isort tool provides stable sorting and grouping, as well as pruning
of unused imports.

Ruff is a modern linter, that is very fast due to being written in rust.
It can detect many common issues in a python codebase.

Removes the pylint dev requirement, since ruff replaces it.

* treewide: fix issues detected by ruff

* treewide: fix bare except clauses

* .devcontainer: Set up isort

* treewide: optimize imports

* treewide: apply black

* treewide: make regex patterns raw strings

This is necessary for escape sequences to be properly recognized.
2023-05-29 05:31:17 -05:00
Blake Blackshear
e10ddb343c
additional shutdown optimizations (#5380) 2023-02-04 08:58:45 -06:00
Blake Blackshear
624c314335
Fast restart (#5378)
* dont wait so long for queues

* implement stop methods for comms

* set the detection events on exit and return early from processing

* handle the stop event in the broadcast threads

* short circuit the detection process exit code if it already exited

* some logging for stats thread

* just keep the log process alive 1 second after the last log message

* ensure the multiprocessing queues are emptied and closed

* Update frigate/log.py

Co-authored-by: Nicolas Mowen <nickmowen213@gmail.com>

* Update frigate/log.py

Co-authored-by: Nicolas Mowen <nickmowen213@gmail.com>

* mypy fixes

---------

Co-authored-by: Nicolas Mowen <nickmowen213@gmail.com>
2023-02-03 20:15:47 -06:00
Dennis George
420bcd7aa0
Convert detectors to factory pattern, ability to set different model for each detector (#4635)
* refactor detectors

* move create_detector and DetectorTypeEnum

* fixed code formatting

* add detector model config models

* fix detector unit tests

* adjust SharedMemory size to largest detector model shape

* fix detector model config defaults

* enable auto-discovery of detectors

* simplify config

* simplify config changes further

* update detectors docs; detect detector configs dynamic

* add suggested changes

* remove custom detector doc

* fix grammar, adjust device defaults
2022-12-15 07:12:52 -06:00
Nate Meyer
e5fe323aca
Add OpenVino Detector (#3768)
* Initial work for adding OpenVino detector.  Not functional

* Load model and submit for inference.

Sucessfully load model and initialize OpenVino engine with either CPU or GPU as device.
Does not parse results for objects.

* Detection working with ssdlite_mobilenetv2 FP16 model

* Add OpenVIno support and model to docker image

* Add documentation for OpenVino detector configuration

* Adds support for ARM32/ARM64 and the Myriad X hardware

-  Use custom-built openvino wheel for all platforms
-  Add libusb build without udev for NCS2 support

* Add documentation around Intel CPU requirements and NCS2 setup

* Print all available output tensors

* Update documentation for config parameters
2022-12-03 10:19:34 -06:00
Nate Meyer
4383b883c0
Refactor to simplify support for additional detector types (#3656)
* Refactor EdgeTPU and CPU model handling to detector submodules.

* Fix selecting the correct detection device type from the config

* Remove detector type check when creating ObjectDetectProcess

* Fixes after rebasing to 0.11

* Add init file to detector folder

* Rename to detect_api

Co-authored-by: Nicolas Mowen <nickmowen213@gmail.com>

* Add unit test for LocalObjectDetector class

* Add configuration for model inputs
Support transforming detection regions to RGB or BGR.
Support specifying the input tensor shape.  The tensor shape has a standard format ["BHWC"] when handed to the detector, but can be transformed in the detector to match the model shape using the model  input_tensor config.

* Add documentation for new model config parameters

* Add input tensor transpose to LocalObjectDetector

* Change the model input tensor config to use an enumeration

* Updates for model config documentation

Co-authored-by: Nicolas Mowen <nickmowen213@gmail.com>
2022-11-03 21:23:09 -05:00
Blake Blackshear
1089a40943 cleanup old code 2020-03-03 20:26:53 -06:00
Blake Blackshear
569e07949f split into separate processes 2020-03-03 20:26:53 -06:00
Blake Blackshear
ebcf1482f8 remove region in process when skipping 2020-02-22 09:03:00 -06:00
Blake Blackshear
50bcad8b77 skip regions when the queue is too full and add more locks 2020-02-22 09:03:00 -06:00
Blake Blackshear
d87f4407a0 switch everything to run off of tracked objects 2020-02-22 09:03:00 -06:00
Blake Blackshear
76c8e3a12f make a copy 2020-02-22 09:03:00 -06:00
Blake Blackshear
10dc56f6ea revamp dockerfile 2020-02-22 09:03:00 -06:00
Blake Blackshear
cc2abe93a6 track objects and add config for tracked objects 2020-02-22 09:03:00 -06:00
Blake Blackshear
f5a2252b29 cleanup imports 2020-02-22 09:03:00 -06:00
Blake Blackshear
5b4c6e50bc dedupe detected objects 2020-02-22 09:03:00 -06:00
Blake Blackshear
9cc46a71cb working dynamic regions, but messy 2020-02-22 09:03:00 -06:00
Blake Blackshear
be1673b00a process detected objects in a queue 2020-02-22 09:03:00 -06:00
Blake Blackshear
b6130e77ff label threads and implements stats endpoint 2020-02-22 09:03:00 -06:00
Blake Blackshear
4180c710cd refactor resizing into generic priority queues 2020-02-22 09:03:00 -06:00
Blake Blackshear
480175d70f add color map to use different colors for different objects 2020-01-02 07:39:57 -06:00
Blake Blackshear
bee99ca6ff track and report all detected object types 2020-01-02 07:39:57 -06:00
blakeblackshear
806acaf445 update dockerignore and debug option 2019-07-30 19:11:22 -05:00
blakeblackshear
baa587028b use a regular subprocess for ffmpeg, refactor bounding box drawing 2019-07-30 19:11:22 -05:00
blakeblackshear
2b51dc3e5b experimental: running ffmpeg directly and capturing raw frames 2019-07-30 19:11:22 -05:00
blakeblackshear
3019b0218c make the threshold configurable per region. fixes #31 2019-05-11 07:39:27 -05:00
blakeblackshear
e0b9b616ce cleanup and update readme 2019-03-30 07:58:31 -05:00
blakeblackshear
edf0cd36df add back flask endpoints 2019-03-29 21:02:40 -05:00
blakeblackshear
0279121d77 WIP: convert to camera class 2019-03-29 20:49:27 -05:00
blakeblackshear
0514eeac03 switch to a thread for object detection 2019-03-27 20:44:57 -05:00
blakeblackshear
a26d2217d4 implement min person size again 2019-03-27 06:45:27 -05:00
blakeblackshear
200d769003 removing motion detection 2019-03-27 06:17:00 -05:00
blakeblackshear
ada8ffccf9 fix for queue size growing too large 2019-03-25 20:35:44 -05:00
blakeblackshear
bca4e78e9a use a queue instead 2019-03-25 06:24:36 -05:00
blakeblackshear
7d3027e056 looping over all regions with motion. ugly, but working 2019-03-20 07:11:38 -05:00
blakeblackshear
c406fda288 fixes 2019-03-19 06:29:58 -05:00
blakeblackshear
8ff9a982b6 start the detection process 2019-03-18 07:48:04 -05:00
blakeblackshear
f2c205be99 prep frames for object detection in a separate process 2019-03-18 07:24:24 -05:00
blakeblackshear
862aa2d3f0 only resize when needed 2019-03-17 20:12:31 -05:00