0 version Please check this article to make sure your GPU is supported. The benefits with this include: Fail-over scenario should a detector not be available (e. yml" and ". I'm getting this when I try to do anyth . CompreFace: Supports both CPU and GPU and is easy to scale up; Is open source and self-hosted, which gives you additional guarantees for data security If you run the GPU version and stop Double Take and manually process images using the CompreFace API does the memory still increase? That will at least point you in the right direction of where it's coming from. 6. then again starts loading subjects into memory. Presume that "compreface-api" should run without crashing and "compreface-core" report being "healthy". Never had working. container_name: "compreface-core". 03) as this was the last driver version to support this GPU. Continuous availability. I saw you opened an issue on CompreFace's GitHub too, I'll follow along with that and see what the outcome is. webcam_demo. How to Challenge Millions of Faces. To use helm you need to install it first, see the instructions here In order to use compreface helm chart you need to configure your own Helm client. thank you for your help in advance … Feb 14, 2023 · Describe the bug Installed Compreface on Debian 11. the default had better recognition. Oct 8, 2021 · maybe i misunderstood, but doubl-take uses "detectores" like compreface for face-detection. " GitHub is where people build software. 0 Multiple gpu support if the machine has 2 or more GPUs, what should I do to make the compreface parallel the work between them? Skip to content Toggle navigation Attaching to compreface-postgres-db, compreface-core, compreface-api, compreface-admin, compreface-ui compreface-admin | Listening for transport dt_socket at address: 5005 compreface-core | [uWSGI] getting INI configuration from uwsgi. it works normally, On this same Host I use deepstack:gpu-2022. SDK supports all functionality from CompreFace. env","contentType Apr 5, 2023 · Add this topic to your repo. Then run CompreFace again docker-compose up -d. I'm getting this when I try to do anything such as uploading a face (using 1. #4482. Author. Contribute to tyabru/CompreFace development by creating an account on GitHub. Apr 9, 2022 · You send logs only from compreface-api node. I did it from scratch, removed postgres docker volume, etc. CompreFace: Supports both CPU and GPU and is easy to scale up; Is open source and self-hosted, which gives you additional guarantees for data security Nov 2, 2021 · i tried the hardest-one without GPU, and the system stood nearly still. The number of matches per second depends on the size of the input image and number of faces on it. md at master · ValentinLevin/CompreFace-1 Subscribe to CompreFace News and Updates to never miss new features and product improvements. The system provides REST API for face recognition, face verification, face 1. 141. jvrobert. 5GB RAM free which is not being used. May 22, 2024 · The system can accurately identify people even when it has only “seen” their photo once. Aug 25, 2023 · Similar for me, I went from compreface:0. Run a benchmark: inside the container docker exec embedding-calculator . (With a ~4GB GPU to reproduce the conditions) Add faces 5 by 5. 108. Building Dec 20, 2021 · double-take won't connect to CompreFace gpu versions despite the Compreface GUI working and processing images. Container name in the docker-compose file: compreface-admin Benchmark. The system provides REST API for face recognition, face verification, face Apr 20, 2023 · b24361c136c7 exadel/compreface-core:1. yml there is an error with compreface-core saying that the base image does not exist. yml and I can indeed choose which GPU to run, but after that, that's it, I can Jun 8, 2023 · I solved my problem, I am using now the compreface-core:1. ini Oct 19, 2021 · Looks like Windows supports GPU if you use WSL2 backend. 0 and 1. env files. You signed out in another tab or window. To increase number of matches per second you can: Compreface build: exadel/compreface:1. You don’t need prior machine learning skills to set up and use CompreFace. now i search for powerfull hardware and GPU. Then delete the volume, e. 1. In case if CompreFace version is older - new SDK features will fail. sh (require exposing API at localhost:3000) Jan 23, 2021 · Saved searches Use saved searches to filter your results more quickly Oct 22, 2022 · Jetson Nano is run on ARM CPU, and compiling CompreFace for ARM is a big challenge. In the default config, it’s also used as the main gateway - Nginx proxies user requests to admin and API servers. Jun 14, 2023 · Hi, for some reason my compreface docker doesn't work anymore. registry - this is the docker hub registry. See this: To Reproduce Steps to reproduce the behavior: Expected behavior Should be able to create an account Screenshots If applicable, add scr The most useful cases include face plugins for face analysis: How to test. When I synchronize the PostgreSQL database with third-party software, facial recognition fails to work correctly. build_version=dev"". 0-arcface-r100-gpu with nvidia GPU, same config I used in older version). g. The ability to use multiple instances of DeepStack (as well as other detectors). g maintenance or timeout). 0-arcface-r100-gpu image with nvidia Quadro M2200 gpu that has only 4gb of memory. Screenshots. Nov 22, 2022 · New User - Coral TPU vs TensorRT GPU. 6、For the configurations compreface_api_java_options=-Xmx8g and compreface_admin_java_options=-Xmx8g, if the memory when running docker-compose for /master/dev/docker-compose-gpu. Anyway, i was hoping there would be a list of servers/minipc's that it would run best on. Just do not forget to update the CompreFace version in config. On the CompreFace application page, at the bottom of the frame, click Create button. Already have an account? In this repository, we have configs and instructions on how to run CompreFace with Kubernetes. so the detector must have an AI-Hardware for face-rec, right? but compreface dont need it, as mentioned in the docs , i found it interesting that it is possible to use GPU (as AI-Hardware) and more precise-facerec models with compreface. Desktop (please complete the following information): OS: Windows 10. Any help would be greatly appreciated. Sep 20, 2022 · Is it enough for server with capability 4x16 without GPU, to achieve 1 second for verifying face 2. I have added UWSGI_PROCESSES: 1 and API_JAVA_OPTS: -Xmx1g env variables in the compose file to limit memory consumption. Aug 18, 2022 · I have installed and successfully launched the Mobilenet-GPU custom build of compreface but for some reason it is not activating uwsgi on the GPU, rather using the CPU (see the logs). Subscribe to CompreFace News and Updates to never miss new features and product improvements. I tried with : docker-compose up -d pospielov/compreface-core:1. Jun 9, 2023 · Starting compreface-fe Starting compreface-fe 2023-06-09 04:59:08,781 INFO spawned: 'compreface-fe' with pid 215 2023-06-09 04:59:08,781 INFO success: compreface-fe entered RUNNING state, process has stayed up for > than 0 seconds (startsecs) compreface-fe: started 2023-06-09 04:59:08,792 INFO exited: startup (exit status 0; expected) Oct 13, 2021 · Saved searches Use saved searches to filter your results more quickly Jun 21, 2021 · In this video, our AI Practice team runs a web camera-based demo application that shows a use case of our face detection serviceCompreFace provides REST API Leading free and open-source face recognition system - Jetson gpu cuda support · exadel-inc/CompreFace@c1bbfc1 I was using an older version of compreface and it worked fine, but decided to move to 1. Scaling Compreface According to the documentation Scalability can be achieved by increasing the number of API servers default = 1 and putting an Embedding Servers in a GPU available instance . Container name in the docker-compose file: compreface-fe. 0". 2, we have added support for Pose Plugin to the UI, making it easier to test. If you want to clear CompreFace installation, first stop it with docker-compose stop . The reason for the 10 seconds detection, was simply that I was using det_prob_threshold=0. Apr 20, 2023 · Interestingly the GPU still has over 1. Jul 20, 2023 · I am running CompreFace in a single container using image: exadel/compreface:1. I've tried DeepStack running on my Jetson Nano 4 gig as well as CompreFace on my Home Assistant Install. Jul 25, 2022 · Features Updated CUDA to 11. 2. CompreFace: Supports both CPU and GPU and is easy to This tutorial shows how to deploy CompreFace using Helm. embedding-calculator && . #1232 opened on Mar 23 by newlife0007. Describe the solution you'd like The system can accurately identify people even when it has only “seen” their photo once. docker volume rm compreface_061_postgres-data. env","contentType Oct 31, 2021 · I haven't used HA, but I see that it has another thing - integrations. For example the SubCenter-ArcFace-r100-gpu build: Original: compreface-core: image: ${registry}compreface-core:${CORE_VERSION} restart: always. I have 3 versions of compreface working each on separate external db's and separate ports (8800,8810,8820). Jul 11, 2023 · Current use of runtime: nvidia in the compreface-core: service of docker-compose. runtime: nvidia. or locally cd . Add support of Pose Plugin on UI. 1 version (Ampere Nvidia GPU generation is supported) Migration from 1. I'm planning to use my server (i5-4690K, RAID, Ubuntu 22. Could you send logs from compreface-core logs? If it's not working then probably you need more RAM. In case if CompreFace version is newer - SDK won't support new features of CompreFace. How can i enable this instead of using the CPU? I've tried following all online docs relating to the WSL2 but when processing images, it is still using the CPU not the GPU. 2. yml for GPU enabled builds means that they will not work on Docker Desktop Windows (At least I have not being able to find a solution on how to install the Nvidia runtime on Windows) Describe the solution you'd like Nov 2, 2021 · as there is nothing else on this server than the container for compreface therefore on the GPU is nothing else running than what is configured inside the compreface-core containter for the nivida runtime. I think adding Google Coral support is easier as soon as it is used on x86 machine. The system can accurately identify people even when it has only “seen” their photo once. Enter the name of the service you are going to create. However, beyond 20 faces, the accuracy noticeably decreases. 0 thinking that it was a sane default, and also using limit=1 , so basically the limit of 1 was masking the problem with setting the threshold to zero. For release and pre-build images, it should be set to exadel/ value. And i am guessing Compreface does not support Coral (ever?) Otherwise you would have release a build/option for it already i guess. Reload to refresh your session. If you use the CPU versi In the release archive and all custom builds, there is a . Template is the GPU arcface 1. Leading free and open-source face recognition system for coral - johnsmoses/CompreFace-tpunet CompreFace is delivered as a docker-compose config and supports different models that work on CPU and GPU. For 5000 faces probably you need about 8-10 Gb of RAM. JavaScript SDK for CompreFace - free and open-source face recognition system from Exadel exadel-inc/compreface-javascript-sdk’s past year of commit activity JavaScript 69 Apache-2. Aug 22, 2023 · Features. This computer has a NVIDIA RTX GPU. Jun 30, 2022 · first, I run docker pull pospielov/compreface-core:1. Setup: OS:Ubuntu 22. In the photo below, only 18 faces were recognized out of 32，use SubCenter-ArcFace-r100-gpu。. Oct 27, 2022 · all little hacks really, but now at least I have a working docker-compose-gpu. However, it is consuming all the available gpu memory. Scaling is needed for increasing the number of requests per second CompreFace can handle. 0 Features Massive UX/UI redesign Wizard for creating the first facial recognition service Pose plugin Service statistics Automatic login after registration Updated session logic, no need to re-login Mar 10, 2024 · The recognition rate is very high in images containing up to 10 faces. /benchmark. env file with configuration options for CompreFace. #1230 opened on Feb 25 by lschapker. 1 and it's working ok, I also use frigate . Features. You switched accounts on another tab or window. postgres_username - username The system can accurately identify people even when it has only “seen” their photo once. CompreFace: Supports both CPU and GPU and is easy to scale up; Is open source and self-hosted, which gives you additional guarantees for data security If you want to clear CompreFace installation, first stop it with docker-compose stop. To associate your repository with the compreface-sdk topic, visit your repo's landing page and select "manage topics. Perform the following steps: Build and run embedding-calculator with the needed scanner backend and CPU/GPU supports. 1. Even without a TPU or GPU, CompreFace consistently scores more correct matches than DeepStack on the Nano. I was using an older version of compreface and it worked fine, but decided to move to 1. Aug 22, 2023 · Exadel CompreFace is a free and open-source face recognition GitHub project. I did manage to build it manually previously, but just wanted a working automated docker-compose file so that upgrades were easier. This container runs Nginx that serves CompreFace UI. however when I run Compreface/SubCenter-ArcFace-r100. Not exactly sure. I am having issues where the CPU usage rockets sky high after running compreface for a few hours. also will it be possible to add 100000 face collections if were to scale. 0-mobilenet-2d106 "I have CompreFace face already installed and running with the default docker image 5 landmarks at the same server using my public IP and everything is working fine" . In CompreFace 1. # detector settings (default: shown below) detectors: compreface: url: # recognition api key key: # number of seconds before the request times out and is aborted timeout: 15 # minimum required confidence that a recognized face is actually a face # value is between 0. But in our R&D sessions, we're seeing an average of 15-20 seconds per image processing time. API key for this demo was created on step 5 of How to Use CompreFace. . To run a demo, open an HTML file in a browser. 2 repository. Memory usage shouldn't grow over time and previously processed images shouldn't mess up with the GPU's memory. Expected behavior. i did the same with the default facenet-model. OS: Unraid Custom Build: Mobilenet-GPU GPU: NVIDIA Quadro P5000 A tag already exists with the provided branch name. That being said I do have to use older drivers (v470. 8 # require opencv to find a face before Leading free and open-source face recognition system for coral - johnsmoses/CompreFace-tpunet compreface. than i changed to mobilenet and retrained one folder. If I change my double-take config to the gpu url and api key it just won't connect and then timeout. Jun 6, 2023 · Saved searches Use saved searches to filter your results more quickly Leading free and open-source face recognition system - exadel-inc/CompreFace Jul 22, 2022 · You signed in with another tab or window. 0-arcface-r100-gpu; Sign up for free to join this conversation on GitHub. To update the CompreFace version or change custom build, download new docker-compose. 0 det_prob_threshold: 0. 🟡 SDK works with this CompreFace version. CompreFace: Supports both CPU and GPU and is easy to scale up; Is open source and self-hosted, which gives you additional guarantees for data security Elytum commented on Jan 22. Discuss code, ask questions & collaborate with the developer community. Images are then processed though Facebox and/or CompreFace for identification. For production systems, we recommend looking through them and set up CompreFace accordingly. Essentially, it is a docker-based application that can be used as a standalone server or deployed in the cloud. I've started it by passing a specific device id in the docker-compose. There are major backward compatibility issues. I would like to add additional DeepStack instances and allow Double-Take to use them. but randomly it releases all the memory. 3. 1-arcface-r100-gpu to compreface:1. Steps to reproduce the behavior: Run CompreFace (ArcFace GPU) in WSL2. Updated CUDA to version 11. CompreFace: Supports both CPU and GPU and is easy to scale up; Is open source and self-hosted, which gives you additional guarantees for data security Nov 2, 2021 · If yes - it should work with GPU with an index 0 To change this - you can set GPU_IDX environment variable But, I am not sure that the problem with compreface-core container. I am running this with Frigate and Double Take. #1231 opened on Mar 10 by songweionline. 0-mobilenet-2d106 but Leading free and open-source face recognition system - Jetson gpu cuda support · exadel-inc/CompreFace@394780f Launch ArcFace GPU with only 1 process. 0 "uwsgi --ini uwsgi. May 18, 2024 · martinenkoEduard asked Jul 8, 2022 in Q&A · Answered. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Jul 11, 2023 · To be able to use a GPU enabled build in Docker Desktop Windows, we have to edit the docker-compose. Consume the API for 4-5h (sometimes less). yml and . Mar 6, 2023 · You signed in with another tab or window. I think the problem can be with compreface-api container. CUDA: Check failed: e == cudaSuccess || e == cudaErrorCudartUnloading: no CUDA-capable device is detected. CompreFace default config doesn't support GPU, but you can use one of the custom configs: Jan 3, 2022 · Please describe. Hi ! For some reason, whenever I start compreface, no matter the values of uwsgi_processes and uwsgi_threads, the processes are all started on the first available GPU. We would like to show you a description here but the site won’t allow us. can you tell me where c Contribute to dgkapanke/compreface development by creating an account on GitHub. Regarding manually pulling images, you can pull single image: docker pull exadel/compreface:1. Explore the GitHub Discussions forum for exadel-inc CompreFace. Features Contribute to eyalfink/CompreFace development by creating an account on GitHub. compreface-api service in constant "crash loop". Currently have it running on a old iMac which seems to work fine (intel etc) but anything with a GPU in it would be even better. The output from the API is an array of matching faces, which I then process through Node-Red to update the location for the matching users in Home Assistant. compreface_061_postgres-data. Feb 27, 2023 · Core occupies memory until eventually taking all memory available. In the Create Service dialog, from the Type drop-down menu, select DETECTION. I am using double-take to get compreface process the images. env","path":"custom-builds/Mobilenet-gpu/. 03 after the docker starts, compreface-api container start populating RAM with the subjects for few minutes, then the recognition works fine. I've already toyed with Frigate by Leading free and open-source face recognition system - CompreFace/Makefile at master · exadel-inc/CompreFace Oct 7, 2022 · CompreFace 1. So my thought was to see if I could put CompreFace on the Nano and that way it can use the CUDA cores for the neural processing. If not , could you please tell what kind of server should be used for compreFace with criteria like ram, storage, cpu architecture and etc. Went to UI URL. 1, we added Pose Plugin, which allows you to detect the pose of a face in an image. More than 100 million people use GitHub to discover, fork, and contribute to over 420 million projects. html; This demo shows the most simple webcam demo for Face recognition service. 8 and added support for the new generation of Nvidia GPUs, including the Ada Lovelace and Hopper microarchitectures. Jul 27, 2022 · GPU Support and information. 0 35 4 13 Updated Sep 25, 2023 Steps to reproduce the behavior: install "stack" using "Portainer" (using the provided "docker-compose. I'm planning on getting three 1080p ip cameras (maybe four). Used Subcenter-arcface-r100 to remove false positives. My server already runs a bunch of docker services, but rarely goes above 5% CPU usage and usually has around 12GB of memory free. could you run docker ps and check if it's running and run docker logs compreface-api CompreFace architecture diagram. i will report back . Balancer + UI. 0-arcface-r100-gpu and many pictures where faces were recognised and identified before now shows as "no face" Sure, many of them are a bit darker, or the angle of the face is not 100% straight forward. Leading free and open-source face recognition system - CompreFace-1/Architecture-and-scalability. To Reproduce. and the recognition also fails. Admin server. yml file of the build. Mar 20, 2023 · machine: x86_64 clock source: unix detected number of CPU cores: 24 current working directory: /app/ml detected binary path: /usr/local/bin/uwsgi !!! no internal routing support, rebuild with pcre support !!! 2023-03-23 13:51:03,846 INFO success: compreface-core entered RUNNING state, process has stayed up for > than 0 seconds (startsecs) setgid() to 33 setuid() to 33 your processes number The system can accurately identify people even when it has only “seen” their photo once. Whenever an images needs to be processed compreface-core shows the captio I bought this GPU based on the fact that it works with compreface 1. Wait a few minutes (~10 minutes in my last test) Expected behavior. In this case, you have to run CompreFace himself, but there is still an integration with HA. To find the volume name, run docker volume ls, the name should be <CompreFace folder>_postgres-data, e. Is it correct that it's using this much memory, and if so what is the minimum requirement to use it? The text was updated successfully, but these errors were encountered: Oct 27, 2022 · Came across compreface because AWS Rekognition has a high cost. I haven't tried it though. {"payload":{"allShortcutsEnabled":false,"fileTree":{"custom-builds/Mobilenet-gpu":{"items":[{"name":". CompreFace is running on Desktop Docker. Support to try the hardest modell and compare again. News and updates. Code Snippets Feb 25, 2021 · The API is exposed with a POST method /recognize which expects the MQTT frigate/events JSON as the payload. env" files from version "v1. 01. yml for an AVX only CPU using facenet and GPU. Exadel CompreFace is a free and open-source face recognition GitHub project. 0. log Add this topic to your repo. 04; GPU: GTX 970; Driver: 510. I'm currently using Windows 10, i7. To associate your repository with the compreface topic, visit your repo's landing page and select "manage topics. Will it work for you? CompreFace addon will be used as an engine by the Double Take addon, right now we are using CompreFace in an unsupported way manually installing it into HA's Sep 14, 2023 · To see the list of GPUs supported by older CompreFace versions, visit our previous blog post. Our solution is based on state-of-the-art methods and libraries like FaceNet and InsightFace. i…" 6 weeks ago Up 3 weeks 3000/tcp compreface-core myDockerCompose. Technology-wise, CompreFace has several advantages over similar free face recognition solutions. 04) with Frigate. You can see from the image attached the CPU usage spike until i stop the compreface container, usage goes up to around 90%. And then running it with Jetson Nano GPU support will be an even bigger challenge. 01 as per the posted compatible GPU's (not that I am trying to place blame - just stating it is supposed to be compatible). 0-arcface-r100 build, no GPU required. tried to create account. cq tc fd uv ra ws zs sv kq tp