YOLOv8旋转对象检测
YOLOv8框架在在支持分类、对象检测、实例分割、姿态评估的基础上更近一步，现已经支持旋转对象检测（OBB），基于DOTA数据集，支持航拍图像的15个类别对象检测，包括车辆、船只、典型各种场地等。包含2800多张图像、18W个实例对象。
YOLO OBB标注数据格式，主要是类别与四个角点归一化到0～1之间的坐标，格式表示如下：
class_index, x1, y1, x2, y2, x3, y3, x4, y4
训练以后的YOLOv8预测xyhwr + 类别数目，不同尺度的YOLOv8 OBB模型的精度与输入格式列表如下：
图片
导出与预测
基于YOLOv8命令行推理测试：
## 导出 yolo export model=yolov8s-obb.pt format=onnx ## 推理 yolo obb predict model=yolov8n-obb.pt source=plane_03.jpg
ONNX推理代码演示
基于OpenVINO2023与ONNX格式模型直接预测推理，首先看一下ONNX格式的YOLOv8-OBB输入与输出格式：
旋转对象检测-代码演示
class_list = load_classes() colors = [(255, 255, 0), (0, 255, 0), (0, 255, 255), (255, 0, 0)] ie = Core() for device in ie.available_devices: print(device) # Read IR model = ie.read_model(model="yolov8s-obb.onnx") compiled_model = ie.compile_model(model=model, device_name="CPU") output_layer = compiled_model.output(0) ## xywhr frame = cv.imread("D:/wh860.jpg") # frame = cv.imread("wh300.jpg") # frame = cv.imread("obb_01.jpeg") bgr = format_yolov8(frame) img_h, img_w, img_c = bgr.shape start = time.time() image = cv.dnn.blobFromImage(bgr, 1 / 255.0, (1024, 1024), swapRB=True, crop=False) res = compiled_model([image])[output_layer] # 1x25x8400 rows = np.squeeze(res, 0).T boxes, confidences, angles, class_ids = post_process(rows) indexes = cv.dnn.NMSBoxes(boxes, confidences, 0.25, 0.45) M = np.zeros((2, 3), dtype=np.float32) for index in indexes: box = boxes[index] d1 = -angles[index] color = colors[int(class_ids[index]) % len(colors)] pts = [(box[0], box[1]), (box[0]+box[2], box[1]), (box[0]+box[2], box[1]+box[3]), (box[0], box[1]+box[3])] rrt_pts = get_rotate_point(pts, M, d1, box) cv.drawContours(frame, [np.asarray(rrt_pts).astype(np.int32)], 0, (255, 0, 255), 2) cv.rectangle(frame, (box[0], box[1] - 20), (box[0] + box[2], box[1]), color, -1) cv.putText(frame, class_list[class_ids[index]], (box[0], box[1]-8), cv.FONT_HERSHEY_SIMPLEX, .5, (255, 255, 255)) end = time.time() inf_end = end - start fps = 1 / inf_end fps_label = "FPS: %.2f" % fps cv.putText(frame, fps_label, (20, 45), cv.FONT_HERSHEY_SIMPLEX, 1, (0, 0, 255), 2) cv.imshow("YOLOv8-obb+OpenVINO2023.x Object Detection", frame) cv.imwrite("D:/wk_result.jpg", frame) cv.waitKey(0) cv.destroyAllWindows()
运行结果如下：
