This requires modifying the code of nvinfer(/opt/nvidia/deepstream/deepstream/sources/gst-plugins/gst-nvinfer).
You can refer this FAQ.
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| how to set scale in nvinfer plugin? | 2 | 1008 | October 12, 2021 | |
| Input Resolution of default deepstream caffe model | 6 | 694 | October 12, 2021 | |
| Bounding box coordinates after nvinfer | 3 | 306 | November 29, 2022 | |
| Whate the picture change from1280x960 to 416x416 between "nvstreammux" and "nvinfer" | 2 | 330 | October 18, 2021 | |
| Getting image from deepstream | 5 | 1450 | October 12, 2021 | |
| Nvinfer input Image padding and scaling | 6 | 1410 | October 12, 2021 | |
| Questions about get_converted_buffer in gstnvinfer.cpp | 5 | 1508 | October 12, 2021 | |
| add nvtracker and some nvinfer in deepstream_test3_app but only one nvinfer work | 2 | 1191 | October 12, 2021 | |
| About the resize method in nvvideoconvert/nvstreammux | 8 | 2188 | October 12, 2021 | |
| cascade original image and resized image before inference | 2 | 615 | October 12, 2021 |