Comments (10)
The last DL-Streamer's release "Release 2023.0" says it was built with
the 2023.0.0 version of the OpenVINO™ Toolkit
(from "https://github.com/dlstreamer/dlstreamer/releases/tag/2023.0-release")
There were no updates to DL-Streamer since then...
If you require the latest OpenVINO then you either need to wait for an DL-Streamer update - or try to patch DL-Streamer on your own (regarding color formats) for now...
from dlstreamer.
we just released 2024.0 version with support of API 2.0. I highly recommend to check it.
from dlstreamer.
The last DL-Streamer's release "Release 2023.0" says it was built with
the 2023.0.0 version of the OpenVINO™ Toolkit
(from "https://github.com/dlstreamer/dlstreamer/releases/tag/2023.0-release")
There were no updates to DL-Streamer since then...If you require the latest OpenVINO then you either need to wait for an DL-Streamer update - or try to patch DL-Streamer on your own (regarding color formats) for now...
Thanks for your comment! Could you please help provide this patch to us? If can't, how difficult do you think if we create this patch by ourselves? And could you point out the potential files regarding color formats?
from dlstreamer.
The two color formats NV12 and I420 were removed from OpenVINO in this PullRequest: "openvinotoolkit/openvino#17919", it is quite a big change in OpenVINO.
Do you really need the latest version of OpenVINO or can you use the previous release of OpenVINO?
from dlstreamer.
Yes, we need latest version of OpenVINO, because for renset50 and mobilenet-v2 we saw better inference throughput on latest OpenVINO vs previous release. we need that improvement.
from dlstreamer.
pls see the perf improvement for resnet50 and mobilenet-v2 on OpenVINO 23.2.
from dlstreamer.
Does your use-case require NV12 and/or I420 color formats, e.g. doing GPU-accelerated video-decoding (normally the Intel-GPU-video-codecs use NV12)? Or do you use picture inputs (which often are variants of RGB formats only)? In latter case you might just start disabling the lines causing compile-errors regarding missing NV12 and I420 enum symbols...
Maybe the DL-Streamer dev-team is already working on a new DL-Streamer release. Maybe there is already something available in a branch or Docker container image?
@akwrobel can you comment, please?
from dlstreamer.
It is first case, we use the iGPU for video decoding(the format of decoded frame is NV12) and then perform inference on iGPU as well. Looks need to add NV12 and/or I420 (API1.0) back in OpenVINO 23.2, or Let DLstreamer to call new API2.0 of OpenVINO.
from dlstreamer.
The DL-Streamer dev-team has already started to port to API2.0 with a new architecture (it was mentioned in another issue). That's why I thought there is already something available in a special branch or an experimental Docker container image available.
It might already be vacation saison (Christmas)... not sure who from the dev-team is still available.
from dlstreamer.
When does the DLstream dev team anticipate completing the port to API 2.0? Will this feature be included in the next release of DLstream?
from dlstreamer.
Related Issues (20)
- Segfault when reuse same model with same model-instance-id. HOT 3
- Installation Error DLStreamer - Both Debian Packages and Compile from Sources HOT 16
- Performance DLStreamer (Gstreamer) vs Pure Openvino (without Gstreamer) for multiple real-time streams
- Docker image for arm64 HOT 2
- How to generate model proc file for yolov8? HOT 2
- yolox support HOT 3
- EfficientNet-B1 support HOT 21
- ERROR: from element /GstPipeline:pipeline0/GstGvaDetect:gvadetect0: base_inference plugin initialization failed HOT 5
- cant run againts my camera feed HOT 46
- does dlstreamer supporst yolov8? HOT 2
- with Docker cmd, cant create and dowload models inside docker HOT 7
- ffmpeg_openvino build failed, LibAV HOT 8
- ffmpeg_openvino/decode_resize_inference build error HOT 9
- cant run against Efficientnet-b0 due to model exceeds allowable size of 10MB HOT 16
- detection with yolo not avaiable on latest dl_streamer docker HOT 33
- cant run 2 dlstreamer cmds side to side on same docker using 2 different camera feeds HOT 23
- there is a customer's detect model need to support HOT 3
- Can we specify the IOU threshold in yolov8 post-process json like yolov5? HOT 19
- Update aligned image for the next classifier HOT 3
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from dlstreamer.