search

jkjung-avt / tensorrt_demos

TensorRT MODNet, YOLOv4, YOLOv3, SSD, MTCNN, and GoogLeNet
https://jkjung-avt.github.io/
MIT License
1.74k stars 545 forks source link

Batch Processed Image Inference #591

Closed loginabhay closed 1 year ago

loginabhay commented 1 year ago

Hi i was able to run the batch inference for yolov4 trt model but after getting the output how to identify which output belongs to which image from the created batch of frames currently getting the output classes as this for 4 images.

Classes: [0. 1. 2. 2. 6. 6. 6. 7. 7. 7. 7.]

How to map these to the images. Kindly help me here

jkjung-avt commented 1 year ago

Please refer to https://github.com/jkjung-avt/tensorrt_demos/issues/306.

Basically, you need to separate inference results for image before you do NMS.

loginabhay commented 1 year ago

@jkjung-avt Thanks for the quick response and amazing work i was able to solve that.