They’d better match the stream resolution in the nvstreammux.
It depends. You can refer to our FAQ to learn how to improve the accuracy.
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| YOLOv2 performance Issue based on streammux output size | 5 | 492 | October 12, 2021 | |
| Nvstreammux enable-padding logic | 10 | 433 | February 5, 2024 | |
| Nvinfer input Image padding and scaling | 6 | 1323 | October 12, 2021 | |
| Explanation about batch-size | 5 | 840 | November 16, 2021 | |
| Python vs C++ Parser Conversion Issues in DeepStream with YOLOv8 | 4 | 495 | January 3, 2024 | |
| Enable padding in right and bottom of input | 2 | 378 | October 12, 2021 | |
| How can I find source code for streammux in Deepstream? | 12 | 351 | September 15, 2023 | |
| Handle streams with different resolutions | 7 | 656 | December 16, 2022 | |
| Inference results different when scaling is executed on NvStreammux | 5 | 56 | February 11, 2025 | |
| Deepstream output has a low quality | 2 | 579 | November 10, 2022 |