debug request

openpose_trt.py

import pycuda.driver as cuda
import pycuda.autoinit
import numpy as np
import tensorrt as trt
import cv2
from hyperpose import Config,Model


TRT_LOGGER = trt.Logger()

# Simple helper data class that's a little nicer to use than a 2-tuple.
class HostDeviceMem(object):
	def __init__(self, host_mem, device_mem):
		self.host = host_mem
		self.device = device_mem

	def __str__(self):
		return "Host:\n" + str(self.host) + "\nDevice:\n" + str(self.device)

	def __repr__(self):
		return self.__str__()

# Allocates all buffers required for an engine, i.e. host/device inputs/outputs.
def allocate_buffers(engine):
	inputs = []
	outputs = []
	bindings = []
	stream = cuda.Stream()
	out_shapes = []
	input_shapes = []
	out_names = []
	max_batch_size = engine.max_batch_size
	for binding in engine:
		binding_shape = engine.get_binding_shape(binding)
		#Fix -1 dimension for proper memory allocation for batch_size > 1
		if binding_shape[0] == -1:
			binding_shape = (1,) + binding_shape[1:]
		size = trt.volume(binding_shape) * max_batch_size
		dtype = trt.nptype(engine.get_binding_dtype(binding))
		# Allocate host and device buffers
		host_mem = cuda.pagelocked_empty(size, dtype)
		device_mem = cuda.mem_alloc(host_mem.nbytes)
		# Append the device buffer to device bindings.
		bindings.append(int(device_mem))
		# Append to the appropriate list.
		if engine.binding_is_input(binding):
			inputs.append(HostDeviceMem(host_mem, device_mem))
			input_shapes.append(engine.get_binding_shape(binding))
		else:
			outputs.append(HostDeviceMem(host_mem, device_mem))
			#Collect original output shapes and names from engine
			out_shapes.append(engine.get_binding_shape(binding))
			out_names.append(binding)
	return inputs, outputs, bindings, stream, input_shapes, out_shapes, out_names, max_batch_size

# This function is generalized for multiple inputs/outputs.
# inputs and outputs are expected to be lists of HostDeviceMem objects.
def do_inference(context, bindings, inputs, outputs, stream):
	# Transfer input data to the GPU.
	[cuda.memcpy_htod_async(inp.device, inp.host, stream) for inp in inputs]
	# Run inference.
	context.execute_async_v2(bindings=bindings, stream_handle=stream.handle)
	# Transfer predictions back from the GPU.
	[cuda.memcpy_dtoh_async(out.host, out.device, stream) for out in outputs]
	# Synchronize the stream
	stream.synchronize()
	# Return only the host outputs.
	return [out.host for out in outputs]

class TrtModel(object):
	def __init__(self, model):
		self.engine_file = model
		self.engine = None
		self.inputs = None
		self.outputs = None
		self.bindings = None
		self.stream = None
		self.context = None
		self.input_shapes = None
		self.out_shapes = None
		self.max_batch_size = 1
		self.cuda_ctx = cuda.Device(0).make_context()
		if self.cuda_ctx:
			self.cuda_ctx.push()


	def build(self):
		
		with open(self.engine_file, 'rb') as f, trt.Runtime(TRT_LOGGER) as runtime:
			self.engine = runtime.deserialize_cuda_engine(f.read())
		self.inputs, self.outputs, self.bindings, self.stream, self.input_shapes, self.out_shapes, self.out_names, self.max_batch_size = allocate_buffers(
			self.engine)

		self.context = self.engine.create_execution_context()
		self.context.active_optimization_profile = 0
		if self.cuda_ctx:
			self.cuda_ctx.pop()

	def run(self, input, deflatten: bool = True, as_dict=False):
		
		# lazy load implementation
		if self.engine is None:
			self.build()
		if self.cuda_ctx:
			self.cuda_ctx.push()

		input = np.asarray(input)
		batch_size = input.shape[0]
		allocate_place = np.prod(input.shape)
		
		self.inputs[0].host[:allocate_place] = input.flatten(order='C').astype(np.float32)
		self.context.set_binding_shape(0, input.shape)
		trt_outputs = do_inference(
			self.context, bindings=self.bindings,
			inputs=self.inputs, outputs=self.outputs, stream=self.stream)

		if self.cuda_ctx:
			self.cuda_ctx.pop()
		#Reshape TRT outputs to original shape instead of flattened array
		if deflatten:
			trt_outputs = [output.reshape(shape) for output, shape in zip(trt_outputs, self.out_shapes)]
		if as_dict:
			return {name: trt_outputs[i] for i, name in enumerate(self.out_names)}

		return trt_outputs
        # return [trt_outputs[0][:batch_size]]

def preprocess(img):
    # img = cv2.resize(img, (368, 656))
    img = cv2.resize(img, (656, 368))

    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
    img = img.transpose((2, 0, 1)).astype(np.float32)
    img /= 255.0
    return img

engine = TrtModel("/data/data/openpose-coco-V2-HW=368x656.onnx_b1_gpu0_fp16.engine")
engine.build()
image = cv2.imread("/data/pose_test.jpg")
trt_input = preprocess(image)

ori_image = trt_input

trt_outputs = engine.run(trt_input)
conf_map, paf_map = trt_outputs
print("trt_output: ", np.array(trt_outputs[0]))

Config.set_model_name("openpose-coco")
Config.set_model_type(Config.MODEL.Openpose)
#get visualize function, which is able to get visualized part and limb heatmap image from inferred heatmaps
visualize=Model.get_visualize(Config.MODEL.Openpose)
vis_parts_heatmap,vis_limbs_heatmap=visualize(ori_image,conf_map[0],paf_map[0],save_tofile=True)

CocoLimb=list(zip([1, 8, 9,  1,  11, 12, 1, 2, 3, 1, 5, 6, 1, 0,  0,  14, 15],
                  [8, 9, 10, 11, 12, 13, 2, 3, 4, 5, 6, 7, 0, 14, 15, 16, 17]))
from enum import Enum
class CocoPart(Enum):
    Nose = 0
    Instance = 1
    RShoulder = 2
    RElbow = 3
    RWrist = 4
    LShoulder = 5
    LElbow = 6
    LWrist = 7
    RHip = 8
    RKnee = 9
    RAnkle = 10
    LHip = 11
    LKnee = 12
    LAnkle = 13
    REye = 14
    LEye = 15
    REar = 16
    LEar = 17

#get postprocess function, which is able to get humans that contains assembled detected parts from inferred heatmaps
PostProcessor=Model.get_postprocessor(Config.MODEL.Openpose)
postprocessor=PostProcessor(parts=CocoPart,limbs=CocoLimb,hin=368,\
    win=656,hout=38,wout=46,colors=None)
humans=postprocessor.process(conf_map[0],paf_map[0], 368, 656)
#draw all detected skeletons
output_img=ori_image.copy()
for human in humans:
    output_img=human.draw_human(output_img)