blakeblackshear.frigate/frigate/util/model.py

358 lines
11 KiB
Python

"""Model Utils"""
import logging
import os
import cv2
import numpy as np
import onnxruntime as ort
from frigate.const import MODEL_CACHE_DIR
logger = logging.getLogger(__name__)
### Post Processing
def post_process_dfine(
tensor_output: np.ndarray, width: int, height: int
) -> np.ndarray:
class_ids = tensor_output[0][tensor_output[2] > 0.4]
boxes = tensor_output[1][tensor_output[2] > 0.4]
scores = tensor_output[2][tensor_output[2] > 0.4]
input_shape = np.array([height, width, height, width])
boxes = np.divide(boxes, input_shape, dtype=np.float32)
indices = cv2.dnn.NMSBoxes(boxes, scores, score_threshold=0.4, nms_threshold=0.4)
detections = np.zeros((20, 6), np.float32)
for i, (bbox, confidence, class_id) in enumerate(
zip(boxes[indices], scores[indices], class_ids[indices])
):
if i == 20:
break
detections[i] = [
class_id,
confidence,
bbox[1],
bbox[0],
bbox[3],
bbox[2],
]
return detections
def post_process_rfdetr(tensor_output: list[np.ndarray, np.ndarray]) -> np.ndarray:
boxes = tensor_output[0]
raw_scores = tensor_output[1]
# apply soft max to scores
exp = np.exp(raw_scores - np.max(raw_scores, axis=-1, keepdims=True))
all_scores = exp / np.sum(exp, axis=-1, keepdims=True)
# get highest scoring class from every detection
scores = np.max(all_scores[0, :, 1:], axis=-1)
labels = np.argmax(all_scores[0, :, 1:], axis=-1)
idxs = scores > 0.4
filtered_boxes = boxes[0, idxs]
filtered_scores = scores[idxs]
filtered_labels = labels[idxs]
# convert boxes from [x_center, y_center, width, height]
x_center, y_center, w, h = (
filtered_boxes[:, 0],
filtered_boxes[:, 1],
filtered_boxes[:, 2],
filtered_boxes[:, 3],
)
x_min = x_center - w / 2
y_min = y_center - h / 2
x_max = x_center + w / 2
y_max = y_center + h / 2
filtered_boxes = np.stack([x_min, y_min, x_max, y_max], axis=-1)
# apply nms
indices = cv2.dnn.NMSBoxes(
filtered_boxes, filtered_scores, score_threshold=0.4, nms_threshold=0.4
)
detections = np.zeros((20, 6), np.float32)
for i, (bbox, confidence, class_id) in enumerate(
zip(filtered_boxes[indices], filtered_scores[indices], filtered_labels[indices])
):
if i == 20:
break
detections[i] = [
class_id,
confidence,
bbox[1],
bbox[0],
bbox[3],
bbox[2],
]
return detections
def __post_process_multipart_yolo(
output_list,
width,
height,
):
anchors = [
[(12, 16), (19, 36), (40, 28)],
[(36, 75), (76, 55), (72, 146)],
[(142, 110), (192, 243), (459, 401)],
]
stride_map = {0: 8, 1: 16, 2: 32}
all_boxes = []
all_scores = []
all_class_ids = []
for i, output in enumerate(output_list):
bs, _, ny, nx = output.shape
stride = stride_map[i]
anchor_set = anchors[i]
num_anchors = len(anchor_set)
output = output.reshape(bs, num_anchors, 85, ny, nx)
output = output.transpose(0, 1, 3, 4, 2)
output = output[0]
for a_idx, (anchor_w, anchor_h) in enumerate(anchor_set):
for y in range(ny):
for x in range(nx):
pred = output[a_idx, y, x]
class_probs = pred[5:]
class_id = np.argmax(class_probs)
class_conf = class_probs[class_id]
conf = class_conf * pred[4]
if conf < 0.4:
continue
dx = pred[0]
dy = pred[1]
dw = pred[2]
dh = pred[3]
bx = ((dx * 2.0 - 0.5) + x) * stride
by = ((dy * 2.0 - 0.5) + y) * stride
bw = ((dw * 2.0) ** 2) * anchor_w
bh = ((dh * 2.0) ** 2) * anchor_h
x1 = max(0, bx - bw / 2) / width
y1 = max(0, by - bh / 2) / height
x2 = min(width, bx + bw / 2) / width
y2 = min(height, by + bh / 2) / height
all_boxes.append([x1, y1, x2, y2])
all_scores.append(conf)
all_class_ids.append(class_id)
formatted_boxes = [
[
int(x1 * width),
int(y1 * height),
int((x2 - x1) * width),
int((y2 - y1) * height),
]
for x1, y1, x2, y2 in all_boxes
]
indices = cv2.dnn.NMSBoxes(
bboxes=formatted_boxes,
scores=all_scores,
score_threshold=0.4,
nms_threshold=0.4,
)
results = np.zeros((20, 6), np.float32)
if len(indices) > 0:
for i, idx in enumerate(indices.flatten()[:20]):
class_id = all_class_ids[idx]
conf = all_scores[idx]
x1, y1, x2, y2 = all_boxes[idx]
results[i] = [class_id, conf, y1, x1, y2, x2]
return np.array(results, dtype=np.float32)
def __post_process_nms_yolo(predictions: np.ndarray, width, height) -> np.ndarray:
predictions = np.squeeze(predictions)
# transpose the output so it has order (inferences, class_ids)
if predictions.shape[0] < predictions.shape[1]:
predictions = predictions.T
scores = np.max(predictions[:, 4:], axis=1)
predictions = predictions[scores > 0.4, :]
scores = scores[scores > 0.4]
class_ids = np.argmax(predictions[:, 4:], axis=1)
# Rescale box
boxes = predictions[:, :4]
input_shape = np.array([width, height, width, height])
boxes = np.divide(boxes, input_shape, dtype=np.float32)
indices = cv2.dnn.NMSBoxes(boxes, scores, score_threshold=0.4, nms_threshold=0.4)
detections = np.zeros((20, 6), np.float32)
for i, (bbox, confidence, class_id) in enumerate(
zip(boxes[indices], scores[indices], class_ids[indices])
):
if i == 20:
break
detections[i] = [
class_id,
confidence,
bbox[1] - bbox[3] / 2,
bbox[0] - bbox[2] / 2,
bbox[1] + bbox[3] / 2,
bbox[0] + bbox[2] / 2,
]
return detections
def post_process_yolo(output: list[np.ndarray], width: int, height: int) -> np.ndarray:
if len(output) > 1:
return __post_process_multipart_yolo(output, width, height)
else:
return __post_process_nms_yolo(output[0], width, height)
def post_process_yolox(
predictions: np.ndarray,
width: int,
height: int,
grids: np.ndarray,
expanded_strides: np.ndarray,
) -> np.ndarray:
predictions[..., :2] = (predictions[..., :2] + grids) * expanded_strides
predictions[..., 2:4] = np.exp(predictions[..., 2:4]) * expanded_strides
# process organized predictions
predictions = predictions[0]
boxes = predictions[:, :4]
scores = predictions[:, 4:5] * predictions[:, 5:]
boxes_xyxy = np.ones_like(boxes)
boxes_xyxy[:, 0] = boxes[:, 0] - boxes[:, 2] / 2
boxes_xyxy[:, 1] = boxes[:, 1] - boxes[:, 3] / 2
boxes_xyxy[:, 2] = boxes[:, 0] + boxes[:, 2] / 2
boxes_xyxy[:, 3] = boxes[:, 1] + boxes[:, 3] / 2
cls_inds = scores.argmax(1)
scores = scores[np.arange(len(cls_inds)), cls_inds]
indices = cv2.dnn.NMSBoxes(
boxes_xyxy, scores, score_threshold=0.4, nms_threshold=0.4
)
detections = np.zeros((20, 6), np.float32)
for i, (bbox, confidence, class_id) in enumerate(
zip(boxes_xyxy[indices], scores[indices], cls_inds[indices])
):
if i == 20:
break
detections[i] = [
class_id,
confidence,
bbox[1] / height,
bbox[0] / width,
bbox[3] / height,
bbox[2] / width,
]
return detections
### ONNX Utilities
def get_ort_providers(
force_cpu: bool = False, device: str = "AUTO", requires_fp16: bool = False
) -> tuple[list[str], list[dict[str, any]]]:
if force_cpu:
return (
["CPUExecutionProvider"],
[
{
"enable_cpu_mem_arena": False,
}
],
)
providers = []
options = []
for provider in ort.get_available_providers():
if provider == "CUDAExecutionProvider":
device_id = 0 if not device.isdigit() else int(device)
providers.append(provider)
options.append(
{
"arena_extend_strategy": "kSameAsRequested",
"device_id": device_id,
}
)
elif provider == "TensorrtExecutionProvider":
# TensorrtExecutionProvider uses too much memory without options to control it
# so it is not enabled by default
if device == "Tensorrt":
os.makedirs(
os.path.join(MODEL_CACHE_DIR, "tensorrt/ort/trt-engines"),
exist_ok=True,
)
device_id = 0 if not device.isdigit() else int(device)
providers.append(provider)
options.append(
{
"device_id": device_id,
"trt_fp16_enable": requires_fp16
and os.environ.get("USE_FP_16", "True") != "False",
"trt_timing_cache_enable": True,
"trt_engine_cache_enable": True,
"trt_timing_cache_path": os.path.join(
MODEL_CACHE_DIR, "tensorrt/ort"
),
"trt_engine_cache_path": os.path.join(
MODEL_CACHE_DIR, "tensorrt/ort/trt-engines"
),
}
)
else:
continue
elif provider == "OpenVINOExecutionProvider":
os.makedirs(os.path.join(MODEL_CACHE_DIR, "openvino/ort"), exist_ok=True)
providers.append(provider)
options.append(
{
"arena_extend_strategy": "kSameAsRequested",
"cache_dir": os.path.join(MODEL_CACHE_DIR, "openvino/ort"),
"device_type": device,
}
)
elif provider == "CPUExecutionProvider":
providers.append(provider)
options.append(
{
"enable_cpu_mem_arena": False,
}
)
else:
providers.append(provider)
options.append({})
return (providers, options)