Ncnn: Feature-request: YOLOv4-tiny (detector)

[AlexeyAB]

Feature-request: YOLOv4-tiny (detector)

• source code: https://github.com/AlexeyAB/darknet
• cfg: https://raw.githubusercontent.com/AlexeyAB/darknet/master/cfg/yolov4-tiny.cfg
• weights: https://github.com/AlexeyAB/darknet/releases/download/darknet_yolo_v4_pre/yolov4-tiny.weights
• discussion: https://github.com/AlexeyAB/darknet/issues/6067

Many other features from Darknet were added previously.
There is required only 1 feature:

1. Add groups= and group_id= to the [route] layer.

[route]
layers=-1
groups=2
group_id=1

So if input is WxHxC, it divides input into 2 groups WxHx(C/2) (there are 2 groups: 0 and 1), and loads the 2nd group_1 WxHx(C/2).

If there are many layers specified in layers= parameter, then this will be done for each of the input layers specified in layer=, then results will be concatenated across channels.

[nihui]

Implemented
THANKS!

[AlexeyAB]

@nihui Great!

• Do you know, why Mali-GPU on CPU Kirin 990 gives lower speed than CPU Kirin 990? (HUAWEI P40)
  https://github.com/AlexeyAB/darknet/issues/6091

• Does NCNN support inference on NPU integrated in Smartphones?

• Can we use batch-inference on NCNN?

• Can we quantize yolov4/tiny to FP16/INT8 ?

[flyboy114]

@nihui @AlexeyAB I test yolov4/tiny which download by AlexeyAB released model,though it can run in The android environment ,but the process result is not right. The same picture cannot be recognized after transformation.