• Hardware Platform (Jetson / GPU) GPU
• DeepStream Version 6.1
• JetPack Version (valid for Jetson only)
• TensorRT Version
• NVIDIA GPU Driver Version (valid for GPU only) 535
• Issue Type( questions, new requirements, bugs) questions
Hi, I am trying to run deepstream-python for action recognition with temporal information (e.g. a list of the latest 16 frames). Is there any sample to config deepstream for this purpose? Thanks.
You can refer to our demo deepstream-pose-classification. There is currently no demo with python.
@yuweiw I am checking DeepStream 3D Action Recognition App — DeepStream 6.3 Release documentation. From the document, it loads libnvds_custom_sequence_preprocess to run preprocessing, which caches a list of N frames.
Is it possible to do this (e.g. load this preprocessing for nvstreammux) in deepstream python?
There is no update from you for a period, assuming this is not an issue anymore.
Hence we are closing this topic. If need further support, please open a new one.
Thanks
No. This is an implementation of the preprocess plugin. You can use the python to create the preprocess plugin, but the custom-tensor-preparation-function need to be implemented by C/C++. You can refer to our demo: sources\apps\sample_apps\deepstream-3d-action-recognition