0.15.0 Release Candidate 1 #16131
Replies: 9 comments 23 replies
-
|
Upgraded from 0.14.1, getting repeated disconnects from Unifi Cameras. CPU seems to spike 10-20% as the cameras repeatedly reconnect. |
Beta Was this translation helpful? Give feedback.
-
|
What hardware are you running on? |
Beta Was this translation helpful? Give feedback.
-
|
Using docker compose |
Beta Was this translation helpful? Give feedback.
-
|
You may need to switch back to ffmpeg 5.0 |
Beta Was this translation helpful? Give feedback.
-
|
Switching to 5.0 did stop the crashes, thank you! Is that a 7.0/hardware issue or a Unfi-streams-are-weird issue? Interestingly, CPU remains high-ish (50-60% total) constantly with 5.0 with semantic search. Guessing that the times when CPU was dropping was when FFMPEG was crashing rather then the other way around. Also, Intel GPU consumption drops on 5.0 from mid-20s to mid-10s. |
Beta Was this translation helpful? Give feedback.
-
|
It seems like there is a weird middle ground for Intel iGPUs right before QSV came out where they don't work well with QSV or VAAPI on ffmpeg 7. You could try qsv on ffmepg 7 and see but in general if 5.0 works like it did on 0.14 best to just keep using that |
Beta Was this translation helpful? Give feedback.
-
|
cant export in timelapse, getting this error, regular exports work, on the RC 1 |
Beta Was this translation helpful? Give feedback.
-
|
Thanks, fixed in #16158 |
Beta Was this translation helpful? Give feedback.
-
|
awesome, thanks |
Beta Was this translation helpful? Give feedback.
-
|
I think the 640x640 makes me want to switch from my Coral to my Nvidia 2060. Since I have Frigate+ and my custom model, is this the only change needed? Can I ignore the instructions from the docs below?
|
Beta Was this translation helpful? Give feedback.
-
In my personal experience it was not the same, yolonas 320 was already better at detecting further away objects. But again your mileage will vary. |
Beta Was this translation helpful? Give feedback.
-
|
@NickM-27 Do you know why the yolonas model performs better, at least in this regard, than the mobiledet one at the same resolution? |
Beta Was this translation helpful? Give feedback.
-
|
It's an entirely different model architecture and a larger model in general, @blakeblackshear can give a more in depth answer |
Beta Was this translation helpful? Give feedback.
-
|
So many reasons. Mobiledet is an SSD architecture built on tensorflow 1.15. YOLO models are just a completely different architecture. If you go read the research papers about how the networks are built you can get a much better sense for just how different their approaches are. |
Beta Was this translation helpful? Give feedback.
-
|
Right, thank you! I've skimmed what I believe are the relevant papers. I have 8 TPUs in my Frigate machine, but going forward, would you say it makes sense to replace them with NVIDIA GPUs to get the most accurate detection? |
Beta Was this translation helpful? Give feedback.
-
|
So now that v0.15.0 is in RC1, when are we getting v0.16.0-beta1? XD All jokes aside, thanks for all the work you guys put into this project for us all! The last couple of versions have seen some incredible improvements to functionality and accuracy, looking forward to seeing what's to come! |
Beta Was this translation helpful? Give feedback.
-
|
Glad it's working well for you! We're just volunteers that enjoy writing code in our spare time :) 0.16 is a long way off. We're glad you're excited, we are too. |
Beta Was this translation helpful? Give feedback.
-
|
I'm seeing a slow but steady increase in RAM usage on the release candidate (semantic search enabled, small model, GenAI using Gemini). Never noticed this before. Instance previously ran at 1.2-2.0 GB. Started rc1 at around 2GB (late on 25th), now up to 5GB in a fairly linear fashion (well, linearly up during the day, flattens at night, so related to events?). CPU usage steady. I stopped and started the container, it dropped back to 1.6GB. Restarted again with semantic search and genai disabled and I'm down to ~1GB. Running on Ubuntu 24.04 via Docker with an old 750ti and coral PCIe+USB. Four Reolink cameras via RTSP (and restreamed locally for detect/record). Will monitor to see if it reoccurs, might take a couple of days to be sure (then again with semantic search etc reenabled). Any ideas I could try in the meantime? Logs are currently clear of errors ~30 mins after restart.. PS - thanks for all your work on this amazing software! |
Beta Was this translation helpful? Give feedback.
-
|
I can confirm it held steady at 1.2-1.8GB (down to 1.5GB 24 hours after Frigate restart) with semantic search and genai disabled. ~20 hours since I re-enabled both, RAM usage is up to 4GB. It dropped from 2.5-2.3GB overnight, but has been on the rise since then. Is this expected behaviour? |
Beta Was this translation helpful? Give feedback.
-
|
Could someone give me a hint? Spent more time not understanding the error, or why I should change the ?PYTHONHOME? in a container than I'm willing to admit. Get the same error even with almost empty config and "fresh" config and media folders. |
Beta Was this translation helpful? Give feedback.
-
|
This is not normal, seems like an error in the docker compose or something |
Beta Was this translation helpful? Give feedback.
-
|
Is it possible to maintain zoom level while scrubbing thru footage? If not, would it be entertained as a feature? If so I can open a feature request issue |
Beta Was this translation helpful? Give feedback.
-
|
It would be very difficult because it is two entirely different components and the zoom is built in to the high resolution video player. It would essentially mean an entire rewrite of the recordings playback viewer |
Beta Was this translation helpful? Give feedback.
-
|
also in general I would say I think that it is not so useful, when scrubbing you usually want to see the activity as a whole to see where an object is coming / going. When zoomed in might make you miss a detail because only being focused on one specific part. The previews are not high enough resolution for zooming to really add any extra detail |
Beta Was this translation helpful? Give feedback.
-
|
I have this strange problem, at least since beta7, but possibly before. When updating the description of a snapshot in Tracked Object Details, I cannot type a lowercase 'a' character. At first I thought it was my keyboard, but it works everywhere else, and I can copy and paste the letter into the text box from a different window. |
Beta Was this translation helpful? Give feedback.
-
|
this has been fixed in the RC |
Beta Was this translation helpful? Give feedback.
-
|
Hmm, I'm currently running RC1 |
Beta Was this translation helpful? Give feedback.
-
|
Other users have confirmed the fix for this in RC1. Can you try incognito mode in case your browser has cached the old code? |
Beta Was this translation helpful? Give feedback.
-
|
That was it. I didn't even think of that! |
Beta Was this translation helpful? Give feedback.
-
|
Great, glad it's working. |
Beta Was this translation helpful? Give feedback.
-
Beta Documentation: https://deploy-preview-13787--frigate-docs.netlify.app/
Images
ghcr.io/blakeblackshear/frigate:0.15.0-rc1ghcr.io/blakeblackshear/frigate:0.15.0-rc1-standard-arm64ghcr.io/blakeblackshear/frigate:0.15.0-rc1-tensorrtghcr.io/blakeblackshear/frigate:0.15.0-rc1-tensorrt-jp4ghcr.io/blakeblackshear/frigate:0.15.0-rc1-tensorrt-jp5ghcr.io/blakeblackshear/frigate:0.15.0-rc1-rkghcr.io/blakeblackshear/frigate:0.15.0-rc1-rocmghcr.io/blakeblackshear/frigate:0.15.0-rc1-h8lChanges since Beta 7
Major Changes for 0.15.0
Breaking Changes
There are several breaking changes in this release, Frigate will attempt to update the configuration automatically. In some cases manual changes may be required. It is always recommended to back up your current config and database before upgrading:
frigate.dbfileshm_sizeis too low then a warning will be printed in the log stating that it needs to be increased.recordconfig has been refactored to allow for direct control of how longalertsanddetectionsare retained. These values will be automatically populated from your current config, but you may want to adjust the values that are set after updating. See the updated docs here and ensure your updated config retains the footage you want it to.hwaccelpreset they are using (preset-vaapimay need to now bepreset-intel-qsv-h264orpreset-intel-qsv-h265) if camera feeds are not functioning correctly after upgrading. This may need to be adjusted on a per-camera basis. If aqsvpreset is not working properly, you may still need to use apreset-vaapior revert to the previous ffmpeg version as described below.path: "5.0"in yourffmpeg:config entry. For example:ffmpegis no longer part of$PATH. In most cases this is handled automatically.execstreams will need to add the full path for ffmpeg which in most cases will be/usr/lib/ffmpeg/7.0/bin/ffmpeg.bin: ffmpegdefined, it needs to be removed.modelconfig under adetectorhas been simplified to justmodel_pathas a string value. This change will be handled automatically through config migration.Explore
The new Explore pane in Frigate 0.15 makes it easy to explore every object tracked by Frigate. It offers a variety of filters and supports keyword and phrase-based text search, searching for similar images, and searching through descriptive text generated by AI models.
The default Explore pane shows a summary of your most recent tracked objects organized by label. Clicking the small arrow icon at the end of the list will bring you to an infinitely scrolling grid view. The grid view can also be set as the default by changing the view type from the Settings button in the top right corner of the pane.
The Explore pane also serves as the new way to submit images to Frigate+. Filters can be applied to only display tracked objects with snapshots that have not been submitted to Frigate+. The left/right arrow keys on the keyboard allow quick navigation between tracked object snapshots when looking at the Tracked Object Details pane from the grid view.
AI/ML Search
Frigate 0.15 introduces two powerful search features: Semantic Search and GenAI Search. Semantic Search can be enabled on its own, while GenAI Search works in addition to Semantic Search.
Semantic Search
Semantic Search uses a CLIP model to generate embeddings (numerical representations of images) for the thumbnails of your tracked objects, enabling searches based on text descriptions or visual similarity. This is all done locally.
For instance, if Frigate detects and tracks a car, you can use similarity search to see other instances where Frigate detected and tracked that same car. You can also quickly search your tracked objects using an "image caption" approach. Searching for "red car driving on a residential street" or "person in a blue shirt walking on the sidewalk at dawn" or even "a person wearing a black t-shirt with the word 'SPORT' on it" will produce some stunning results.
Semantic Search works by running an AI model locally on your system. Small or underpowered systems like a Raspberry Pi will not run Semantic Search reliably or at all. A dedicated GPU and 16GB of RAM is recommended for best performance.
See the Semantic Search docs for system requirements, setup instructions, and usage tips.
Generative AI
GenAI Search employs generative AI models to create descriptive text for the thumbnails of your tracked objects, which are stored in the Frigate database to enhance future searches. Supported providers include Google Gemini, Ollama, and OpenAI, so you can choose whether you want to send data to the cloud or use a locally hosted provider.
See the GenAI docs for setup instructions and use case suggestions.
Improved Tools for Debugging
Review Item Details Pane
A new Review Item Details pane can be viewed by clicking / tapping on the gray chip on a review item in the Review pane. This shows more information about the review item as well as thumbnails or snapshots for individual objects (if enabled). The pane also provides links to share the review item, download it, submit images to Frigate+, view object lifecycles, and more.
Object Lifecycle Pane
The Recordings Timeline from Frigate 0.13 has been improved upon and returns to 0.15 as the
Object Lifecycle, viewable in the Review Details pane as well as the new Explore page. The new pane shows the significant moments during the object's lifecycle: when it was first seen, when it entered a zone, became stationary, etc. It also provides information about the object's area and size ratio to assist in configuring Frigate to tune out false positives.Native Notifications
Frigate now supports notifications using the WebPush protocol. This allows Frigate to deliver notifications to devices that have registered to receive notifications in the Frigate settings, delivering them in a timely and secure manner. Currently, notifications will be delivered for all review items marked as alerts. More options for native notifications will be supported in the future.
See the notifications docs.
New Object Detectors
ONNX
ONNX is an open model standard which allows for a single model format that can run on different types of GPUs. The
default,tensorrt, androcmFrigate build variants include GPU support for efficient object detection via ONNX models, simplifying configuration and support more models. There are no default included ONNX models for object detection.AMD MiGraphX
Support has been added for AMD GPUs via ROCm and MiGraphX. Currently there is no default included model for this detector.
Hailo-8
Support has been added for the Hailo8 and Hailo-8L hardware for object detection on both arm64 and amd64 platforms.
Other UI Changes
Other Backend Changes
This discussion was created from the release 0.15.0 Release Candidate 1.
Beta Was this translation helpful? Give feedback.
All reactions