display.py

import numpy as np
import cv2
import albumentations as A
from utils import *
import random
from albumentations.pytorch import ToTensorV2
from yolov3 import YOLOV3_PL
from pytorch_grad_cam.utils.image import show_cam_on_image
from utils import YoloCAM, cells_to_bboxes, non_max_suppression
from model import YOLOv3

def inference(image: np.ndarray, iou_thresh: float = 0.5, thresh: float = 0.5,  transparency: float = 0.5):
    model = YOLOV3_PL() #YOLOv3(num_classes=20)
    model.load_state_dict(torch.load("model.pth", map_location=torch.device('cpu')), strict=False)
    # iou_thresh = 0.75
    # thresh = 0.75
    scaled_anchors  =  config.SCALED_ANCHORS
    
   # target_layer_list = model.layers[-2]

   # cam = YoloCAM(model=model, target_layers = target_layer_list, use_cuda=False)
    transforms =  A.Compose(
    [
        A.LongestMaxSize(max_size=config.IMAGE_SIZE),
        A.PadIfNeeded(
            min_height=config.IMAGE_SIZE, min_width=config.IMAGE_SIZE, border_mode=cv2.BORDER_CONSTANT
        ),
        A.Normalize(mean=[0, 0, 0], std=[1, 1, 1], max_pixel_value=255,),
        ToTensorV2(),
    ],
    )
    with torch.no_grad():
        transformed_image = transforms(image=image)["image"].unsqueeze(0)
        output = model(transformed_image)

        bboxes = [[] for _ in range(1)]
        for i in range(3):
            batch_size, A1, S, _, _ = output[i].shape
            anchor = scaled_anchors[i].to('cpu')
            boxes_scale_i = cells_to_bboxes(
                output[i].to('cpu'), anchor, S=S, is_preds=True
            )
            for idx, (box) in enumerate(boxes_scale_i):
                bboxes[idx] += box

    nms_boxes = non_max_suppression(
        bboxes[0], iou_threshold=iou_thresh, threshold=thresh, box_format="midpoint",
    )
    plot_img = draw_predictions(image, nms_boxes, class_labels=config.PASCAL_CLASSES)
    # if not show_cam:
    #     return [plot_img]

    # grayscale_cam = cam(transformed_image, scaled_anchors)[0, :, :]
    # img = cv2.resize(image, (416, 416))
    # img = np.float32(img) / 255
    # cam_image = show_cam_on_image(img, grayscale_cam, use_rgb=True, image_weight=transparency)
    # return [plot_img, cam_image]

    return [plot_img]



def draw_predictions(image: np.ndarray, boxes: list[list], class_labels: list[str]) -> np.ndarray:
    """Plots predicted bounding boxes on the image"""

    colors = [[random.randint(0, 255) for _ in range(3)] for name in class_labels]

    im = np.array(image)
    height, width, _ = im.shape
    bbox_thick = int((height + width) /500)

    # Create a Rectangle patch
    for box in boxes:
        assert len(box) == 6, "box should contain class pred, confidence, x, y, width, height"
        class_pred = box[0]
        conf = box[1]
        box = box[2:]
        upper_left_x = box[0] - box[2] / 2
        upper_left_y = box[1] - box[3] / 2

        x1  = int(upper_left_x * width)
        y1 = int(upper_left_y * height)

        x2 = x1 + int(box[2] * width)
        y2 = y1 + int(box[3] * height)

        cv2.rectangle(
            image,
            (x1, y1), (x2, y2),
            color=colors[int(class_pred)],
            thickness=bbox_thick
        )
        text = f"{class_labels[int(class_pred)]}: {conf:.2f}"
        t_size = cv2.getTextSize(text, 0, 0.7, thickness=bbox_thick // 2)[0]
        c3 = (x1 + t_size[0], y1 - t_size[1] - 3)

        cv2.rectangle(image, (x1, y1), c3, colors[int(class_pred)], -1)
        cv2.putText(
            image,
            text,
            (x1, y1 - 2),
            cv2.FONT_HERSHEY_SIMPLEX,
            0.7,
            (0, 0, 0),
            bbox_thick // 2,
            lineType=cv2.LINE_AA,
        )

    return image