Hi there,
Could you please guide me on how I should be using inference of multiple frames if I am using more than one model for a webcam input? Would I need to use multiple k_LayersPerFrame variables as presented in Split Inference over multiple frames guide?
Thanks.
Yes. You need to decide over how many frames you want to run the model, and compute how many layers per frame. The script from DepthEstimationSample uses a webcam and the model is scheduled over multiple frames.
Hi @montplaisir thanks for responding but I already know how to run it for a single model. My question was regarding splitting inference over multiple frames for more than one model.
Create one Worker for every model, and call ScheduleIterable on each worker. If you want to run all models over the same number of frames it should be a small change from running inference for a single model.