• Hardware Platform Jetson • DeepStream Version 6.3 • JetPack Version 5.1 • TensorRT Version 8.5 • Issue Type questions Hi, my Deepstream application (Yolo as pgie, LPRNet as sgie) performs inference on frames from RTSP stream, and under certain conditions they will be uploaded. Sometimes, the …

[image] nikola2: Is hardware decoding of H.264 and H.265 supported on Jetson Orin AGX 64GB? How can we confirm that hardware decoding/encoding is being used in our DeepStream pipeline? Yes. You can use jtop tool to check the hardware decoding/encoding is being used like below. The NVENC & NV…

Hi, Could you upload an example image showing the artifacts so we can better understand the issue? A few possible causes come to mind: Encoder errors GStreamer freeing the buffer before it’s converted Color space conversion issues One potential solution is to use GStreamer to save the image loc…

Hi, here’s the image with artifacts. [img] Do you have any idea what could it be?

Please refer to our DS_troubleshooting : Solution 11 .

Those definitely look like artifacts from dropped packets or insufficient bitrate in the stream.

So I added this line of code on nvurisrcbin: if "source" in name: source_element = child_proxy.get_by_name("source") if source_element.find_property('drop-on-latency') is not None: Object.set_property("drop-on-latency", False) # set to False to avoid image artifacts…

From your graph, there is no paramters for the source bin and the sink plugin is fakesink. What specific plugins are you using for your source plugin and sink plugin in the previous pipeline?

For source I’m using nvstreammux with nvurisrcbin: if is_live: logger.info("At least one of the sources is live") streammux.set_property('live-source', 1) streammux.set_property('width', 1920) streammux.set_property('height', 1080) streammux.set_property('batch-size…

Can you try performing JPEG encoding using GStreamer? Just add nvvideoconvert and jpegenc before the fakesink, and move the probe accordingly. Also, try saving the buffer you’re receiving before the nvstreammux to see if it already has the artifacts.

Correct inference on frames, but saved frames have artifacts

Accelerated Computing Intelligent Video Analytics DeepStream SDK

yuweiw June 11, 2025, 10:34am 8

Could you refer to our FAQ to get the graph of your pipeline? We can check your configured parameters through this.

Topic		Replies	Views
Help with artefacts needed DeepStream SDK	9	992	February 13, 2023
Saving stream source image results in a black image DeepStream SDK	7	445	January 30, 2024
Intermittent Artifacts in DeepStream RTSP Output with Dynamic Multi-Stream Video Analytics with triton inference server with python backend DeepStream SDK gstreamer , inference-server-triton , deepstream	88	1237	July 8, 2025
Frame rate drops when saving jpg files in Deepstream 6.2 SDK DeepStream SDK	39	2752	April 6, 2023
Pixelated Frames in Deepstream Pipeline DeepStream SDK	20	514	September 30, 2024
DS-7 In ARM , The saved images are all black DeepStream SDK deepstream	31	470	December 5, 2024
Saving video streams DeepStream SDK	2	1155	October 12, 2021
Image taken via tiler_sink_pad_probe is damaged DeepStream SDK	7	331	June 4, 2024
How to save the complete frames and get rtsp out within x_86 system docker container DeepStream SDK	7	711	December 15, 2022
Deeepstream-test5 error when save image DeepStream SDK	35	3184	October 12, 2021

Correct inference on frames, but saved frames have artifacts

Related topics