pil image as input

app.py

@@ -5,11 +5,12 @@ import torch
 from util import Detection
 import os
 
-face_model = os.environ.get('FACE_MODEL')
-age_model = os.environ.get('AGE_MODEL')
+if os.environ.get('FACE_MODEL') is not None:
+    face_model = os.environ.get('FACE_MODEL')
+    age_model = os.environ.get('AGE_MODEL')
 
-torch.hub.download_url_to_file(face_model, 'face_model.pt')
-torch.hub.download_url_to_file(age_model, 'age_model.pt')
+    torch.hub.download_url_to_file(face_model, 'face_model.pt')
+    torch.hub.download_url_to_file(age_model, 'age_model.pt')
 
 sys.path.append("./")
 sys.path.append("./yolov5")
@@ -29,41 +30,39 @@ def run_yolo(img):
 
     img0 = ImageOps.contain(img0, (720,720))
     img0 = ImageOps.exif_transpose(img0)
-    
-    img0.save("img.jpg")
 
     draw = ImageDraw.Draw(img0)
 
-    predictions = predict(age_model_ts, model, stride, names, pt, jit, onnx, engine, imgsz=[320, 320], conf_thres=0.5, iou_thres=0.45, save_conf=True,
-                    exist_ok=True, nosave=True, save_txt=False, source="img.jpg", project=None, name=None)
+    predictions = predict(age_model_ts, model, 
+        stride, imgsz=[320, 320], 
+        conf_thres=0.5, iou_thres=0.45,
+        source=img0
+    )
 
     detections : list[Detection] = []
-    for k, (bboxes, img)  in enumerate(predictions):
-        
-        for i, bbox in enumerate(bboxes):
-            det = Detection(
-                (k+1)*(i+1),
-                bbox["xmin"],
-                bbox["ymin"],
-                bbox["xmax"],
-                bbox["ymax"],
-                bbox["conf"],
-                bbox["class"],
-                bbox["class"],
-                img0.size
-            )
-            same = list(filter(lambda x: x.xmin == det.xmin and x.ymin == det.ymin or ( det.xmin > x.xmin and det.ymin > x.ymin and det.xmax < x.xmax and det.ymax < x.ymax ) or ( det.xmin < x.xmin and det.ymin < x.ymin and det.xmax > x.xmax and det.ymax > x.ymax ) or Detection.get_iou(det, x) > 0.6, detections))
-
-            if len(same) == 0:
-                detections.append(det)
-                draw.rectangle(((det.xmin, det.ymin), (det.xmax, det.ymax)), fill=None, outline=(255,255,255))
-                draw.rectangle(((det.xmin, det.ymin - 10), (det.xmax, det.ymin)), fill=(255,255,255))
-                draw.text((det.xmin, det.ymin - 10), det.class_name, fill=(0,0,0), font=roboto_font)
-
-    os.remove("img.jpg")
+    for k, bbox  in enumerate(predictions):
+        det = Detection(
+            (k+1),
+            bbox["xmin"],
+            bbox["ymin"],
+            bbox["xmax"],
+            bbox["ymax"],
+            bbox["conf"],
+            bbox["class"],
+            bbox["class"],
+            img0.size
+        )
+
+        detections.append(det)
+        draw.rectangle(((det.xmin, det.ymin), (det.xmax, det.ymax)), fill=None, outline=(255,255,255))
+        draw.rectangle(((det.xmin, det.ymin - 10), (det.xmax, det.ymin)), fill=(255,255,255))
+        draw.text((det.xmin, det.ymin - 10), det.class_name, fill=(0,0,0), font=roboto_font)
+
     return img0
 
 
+# run_yolo("D:\\Download\\IMG_20220803_153335.jpg")
+
 inputs = gr.inputs.Image(type='filepath', label="Input Image")
 outputs = gr.outputs.Image(type="pil", label="Output Image")
 

yolov5/detect.py

@@ -18,7 +18,7 @@ from yolov5.utils.torch_utils import select_device, time_sync
 from yolov5.utils.plots import Annotator, colors, save_one_box
 from yolov5.utils.general import (check_img_size,
                         increment_path, non_max_suppression, scale_coords, xyxy2xywh)
-from yolov5.utils.datasets import IMG_FORMATS, VID_FORMATS, LoadImages
+from yolov5.utils.datasets import IMG_FORMATS, VID_FORMATS, LoadImages, pil_to_cv
 from yolov5.models.common import DetectMultiBackend
 import torchvision
 
@@ -29,188 +29,6 @@ test_transforms = torchvision.transforms.Compose([
     torchvision.transforms.Resize((224, 224)),
 ])
 
-def run(weights=ROOT / 'yolov5s.pt',  # model.pt path(s)
-        source=ROOT / 'data/images',  # file/dir/URL/glob, 0 for webcam
-        data=ROOT / 'data/coco128.yaml',  # dataset.yaml path
-        imgsz=(640, 640),  # inference size (height, width)
-        conf_thres=0.25,  # confidence threshold
-        iou_thres=0.45,  # NMS IOU threshold
-        max_det=1000,  # maximum detections per image
-        device='',  # cuda device, i.e. 0 or 0,1,2,3 or cpu
-        save_img = False,
-        view_img=False,  # show results
-        save_txt=False,  # save results to *.txt
-        save_conf=False,  # save confidences in --save-txt labels
-        save_crop=False,  # save cropped prediction boxes
-        nosave=False,  # do not save images/videos
-        classes=None,  # filter by class: --class 0, or --class 0 2 3
-        agnostic_nms=False,  # class-agnostic NMS
-        augment=False,  # augmented inference
-        visualize=False,  # visualize features
-        update=False,  # update all models
-        project=ROOT / 'runs/detect',  # save results to project/name
-        name='exp',  # save results to project/name
-        exist_ok=False,  # existing project/name ok, do not increment
-        line_thickness=3,  # bounding box thickness (pixels)
-        hide_labels=False,  # hide labels
-        hide_conf=False,  # hide confidences
-        half=False,  # use FP16 half-precision inference
-        dnn=False,  # use OpenCV DNN for ONNX inference
-        ):
-    
-    import torch
-    from utils.torch_utils import select_device, time_sync
-    from utils.plots import Annotator, colors, save_one_box
-    from utils.general import (check_img_size,
-                            increment_path, non_max_suppression, scale_coords, xyxy2xywh)
-    from utils.datasets import IMG_FORMATS, VID_FORMATS, LoadImages
-    from models.common import DetectMultiBackend
-    source = str(source)
-
-    save_dir = None
-    save_path = None
-    # save_img = not nosave and not source.endswith('.txt')  # save inference images
-    is_file = Path(source).suffix[1:] in (IMG_FORMATS + VID_FORMATS)
-    is_url = source.lower().startswith(('rtsp://', 'rtmp://', 'http://', 'https://'))
-    webcam = source.isnumeric() or source.endswith('.txt') or (is_url and not is_file)
-    
-    # Directories
-    if project is not None:
-        save_dir = increment_path(Path(project) / name, exist_ok=exist_ok)  # increment run
-        (save_dir / 'labels' if save_txt else save_dir).mkdir(parents=True, exist_ok=True)  # make dir
-
-    # Load model
-    device = select_device(device)
-    model = DetectMultiBackend(weights, device=device, dnn=dnn, data=data)
-    stride, names, pt, jit, onnx, engine = model.stride, model.names, model.pt, model.jit, model.onnx, model.engine
-    imgsz = check_img_size(imgsz, s=stride)  # check image size
-
-    # Half
-    half &= (pt or jit or onnx or engine) and device.type != 'cpu'  # FP16 supported on limited backends with CUDA
-    if pt or jit:
-        model.model.half() if half else model.model.float()
-
-    dataset = LoadImages(source, img_size=imgsz, stride=stride, auto=pt)
-    bs = 1  # batch_size
-    vid_path, vid_writer = [None] * bs, [None] * bs
-
-    # Run inference
-    model.warmup(imgsz=(1, 3, *imgsz), half=half)  # warmup
-    dt, seen = [0.0, 0.0, 0.0], 0
-
-    #with tqdm(dataset) as pbar:
-    #    pbar.set_description("Document Image Analysis")
-    for path, im, im0s, vid_cap, s in dataset:
-        #print(path)
-        t1 = time_sync()
-        im = torch.from_numpy(im).to(device)
-        im = im.half() if half else im.float()  # uint8 to fp16/32
-        im /= 255  # 0 - 255 to 0.0 - 1.0
-        if len(im.shape) == 3:
-            im = im[None]  # expand for batch dim
-        t2 = time_sync()
-        dt[0] += t2 - t1
-
-        # Inference
-        visualize = increment_path(save_dir / Path(path).stem, mkdir=True) if visualize else False
-        pred = model(im, augment=augment, visualize=visualize)
-        t3 = time_sync()
-        dt[1] += t3 - t2
-
-        # NMS
-        pred = non_max_suppression(pred, conf_thres, iou_thres, classes, agnostic_nms, max_det=max_det)
-        dt[2] += time_sync() - t3
-
-        # Second-stage classifier (optional)
-        # pred = utils.general.apply_classifier(pred, classifier_model, im, im0s)
-
-        # Process predictions
-        preds = []
-        for i, det in enumerate(pred):  # per image
-            seen += 1
-            if webcam:  # batch_size >= 1
-                p, im0, frame = path[i], im0s[i].copy(), dataset.count
-                s += f'{i}: '
-            else:
-                p, im0, frame = path, im0s.copy(), getattr(dataset, 'frame', 0)
-
-            p = Path(p)  # to Path
-            if save_dir is not None:
-                save_path = str(save_dir / p.name)  # im.jpg
-                txt_path = str(save_dir / 'labels' / p.stem) + ('' if dataset.mode == 'image' else f'_{frame}')  # im.txt
-            s += '%gx%g ' % im.shape[2:]  # print string
-            gn = torch.tensor(im0.shape)[[1, 0, 1, 0]]  # normalization gain whwh
-            imc = im0.copy() if save_crop else im0  # for save_crop
-            annotator = Annotator(im0, line_width=line_thickness, example=str(names))
-            if len(det):
-                # Rescale boxes from img_size to im0 size
-                det[:, :4] = scale_coords(im.shape[2:], det[:, :4], im0.shape).round()
-
-                # Print results
-                for c in det[:, -1].unique():
-                    n = (det[:, -1] == c).sum()  # detections per class
-                    s += f"{n} {names[int(c)]}{'s' * (n > 1)}, "  # add to string
-
-                # Write results
-                if save_txt:
-                    with open(txt_path + '.txt', 'w') as f:
-                        for *xyxy, conf, cls in reversed(det):
-                            xywh = (xyxy2xywh(torch.tensor(xyxy).view(1, 4)) / gn).view(-1).tolist()  # normalized xywh
-                            preds.append({"class": str(int(cls)), "xmin": int(xyxy[0]), "ymin": int(xyxy[1]), "xmax": int(xyxy[2]),"ymax": int(xyxy[3]), "conf": float(conf)})
-                            if save_txt:  # Write to file
-                                line = (int(cls), *xywh, conf) if save_conf else (cls, *xywh)  # label format
-                                f.write(('%g ' * len(line)).rstrip() % line + '\n')
-                                    
-                            if save_img or save_crop or view_img:  # Add bbox to image
-                                c = int(cls)  # integer class
-                                label = None if hide_labels else (names[c] if hide_conf else f'{names[c]} {conf:.2f}')
-                                annotator.box_label(xyxy, label, color=colors(c, True))
-                                if save_crop:
-                                    save_one_box(xyxy, imc, file=save_dir / 'crops' / names[c] / f'{p.stem}.jpg', BGR=True)
-                else:
-                    for *xyxy, conf, cls in reversed(det):
-                        xywh = (xyxy2xywh(torch.tensor(xyxy).view(1, 4)) / gn).view(-1).tolist()  # normalized xywh
-                        preds.append({"class": str(int(cls)), "xmin": int(xyxy[0]), "ymin": int(xyxy[1]), "xmax": int(xyxy[2]),"ymax": int(xyxy[3]), "conf": float(conf)})
-                        
-            # Print time (inference-only)
-            # LOGGER.info(f'{s}Done. ({t3 - t2:.3f}s)')
-
-            # Stream results
-            if save_img:
-                im0 = annotator.result()
-                if view_img:
-                    cv2.imshow(str(p), im0)
-                    cv2.waitKey(1)  # 1 millisecond
-
-            # Save results (image with detections)
-            if save_img:
-                if dataset.mode == 'image':
-                    cv2.imwrite(save_path, im0)
-                else:  # 'video' or 'stream'
-                    if vid_path[i] != save_path:  # new video
-                        vid_path[i] = save_path
-                        if isinstance(vid_writer[i], cv2.VideoWriter):
-                            vid_writer[i].release()  # release previous video writer
-                        if vid_cap:  # video
-                            fps = vid_cap.get(cv2.CAP_PROP_FPS)
-                            w = int(vid_cap.get(cv2.CAP_PROP_FRAME_WIDTH))
-                            h = int(vid_cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
-                        else:  # stream
-                            fps, w, h = 30, im0.shape[1], im0.shape[0]
-                        save_path = str(Path(save_path).with_suffix('.mp4'))  # force *.mp4 suffix on results videos
-                        vid_writer[i] = cv2.VideoWriter(save_path, cv2.VideoWriter_fourcc(*'mp4v'), fps, (w, h))
-                    vid_writer[i].write(im0)
-        yield preds, save_path
-    # Print results
-    #t = tuple(x / seen * 1E3 for x in dt)  # speeds per image
-    #LOGGER.info(f'Speed: %.1fms pre-process, %.1fms inference, %.1fms NMS per image at shape {(1, 3, *imgsz)}' % t)
-    """ if save_txt or save_img:
-        s = f"\n{len(list(save_dir.glob('labels/*.txt')))} labels saved to {save_dir / 'labels'}" if save_txt else ''
-        LOGGER.info(f"Results saved to {colorstr('bold', save_dir)}{s}")
-    if update:
-        strip_optimizer(weights)  # update model (to fix SourceChangeWarning) """
-
-
 def load_yolo_model(weights, device="cpu", imgsz=[1280, 1280]):
     # Load model
     device = select_device(device)
@@ -232,90 +50,55 @@ def predict(
 
         age_model,
         model,  # model.pt path(s)
-        stride, names, pt, jit, onnx, engine,
-        source=ROOT / 'data/images',  # file/dir/URL/glob, 0 for webcam
-        data=ROOT / 'data/coco128.yaml',  # dataset.yaml path
+        stride, 
+        source=None,  # PIL Image
         imgsz=(640, 640),  # inference size (height, width)
         conf_thres=0.5,  # confidence threshold
         iou_thres=0.45,  # NMS IOU threshold
         max_det=1000,  # maximum detections per image
         device='cpu',  # cuda device, i.e. 0 or 0,1,2,3 or cpu
-        save_img = False,
-        view_img=False,  # show results
-        save_txt=False,  # save results to *.txt
-        save_conf=False,  # save confidences in --save-txt labels
-        save_crop=False,  # save cropped prediction boxes
-        nosave=False,  # do not save images/videos
         classes=None,  # filter by class: --class 0, or --class 0 2 3
         agnostic_nms=False,  # class-agnostic NMS
         augment=False,  # augmented inference
         visualize=False,  # visualize features
-        update=False,  # update all models
-        project=None,  # save results to project/name
-        name='exp',  # save results to project/name
-        exist_ok=False,  # existing project/name ok, do not increment
-        line_thickness=3,  # bounding box thickness (pixels)
-        hide_labels=False,  # hide labels
-        hide_conf=False,  # hide confidences
         half=False,  # use FP16 half-precision inference
-        dnn=False,  # use OpenCV DNN for ONNX inference
         
         ):
-    
-    
-    source = str(source)
-
-    save_dir = None
-    save_path = None
 
-    dataset = LoadImages(source, img_size=imgsz, stride=stride, auto=pt)
+    im, im0 = pil_to_cv(source, img_size=imgsz[0], stride=stride)
     
-    # Run inference
-
-    dt, seen = [0.0, 0.0, 0.0], 0
-
-    #with tqdm(dataset) as pbar:
-    #    pbar.set_description("Document Image Analysis")
-    for path, im, im0s, vid_cap, s in dataset:
-        #print(path)
-        t1 = time_sync()
-        im = torch.from_numpy(im).to(device)
-        im = im.half() if half else im.float()  # uint8 to fp16/32
-        im /= 255  # 0 - 255 to 0.0 - 1.0
-        if len(im.shape) == 3:
-            im = im[None]  # expand for batch dim
-        t2 = time_sync()
-        dt[0] += t2 - t1
-
-        # Inference
-        visualize = False
-        pred = model(im, augment=augment, visualize=visualize)
-        t3 = time_sync()
-        dt[1] += t3 - t2
-
-        # NMS
-        pred = non_max_suppression(pred, conf_thres, iou_thres, classes, agnostic_nms, max_det=max_det)
-        dt[2] += time_sync() - t3
-
-        # Process predictions
-        preds = []
-        for i, det in enumerate(pred):  # per image
-            
-            p, im0, frame = path, im0s.copy(), getattr(dataset, 'frame', 0)
-            
-            if len(det):
-                # Rescale boxes from img_size to im0 size
-                det[:, :4] = scale_coords(im.shape[2:], det[:, :4], im0.shape).round()
-
-                for *xyxy, conf, cls in reversed(det):
-                    face = im0[int(xyxy[1]):int(xyxy[3]),int(xyxy[0]):int(xyxy[2])]
-                    face_img = cv2.cvtColor(face, cv2.COLOR_BGR2RGB)
-                    im = test_transforms(face_img).unsqueeze_(0)
-                    with torch.no_grad():
-                        y = age_model(im)
-                    
-                    age = y[0]
-
-                    preds.append({"class": str(int(age)), "xmin": int(xyxy[0]), "ymin": int(xyxy[1]), "xmax": int(xyxy[2]),"ymax": int(xyxy[3]), "conf": float(conf)})
-                    
-        yield preds, save_path
+    im = torch.from_numpy(im).to(device)
+    im = im.half() if half else im.float()  # uint8 to fp16/32
+    im /= 255  # 0 - 255 to 0.0 - 1.0
+    if len(im.shape) == 3:
+        im = im[None]  # expand for batch dim
+   
+    # Inference
+    visualize = False
+    pred = model(im, augment=augment, visualize=visualize)
+
+    # NMS
+    pred = non_max_suppression(pred, conf_thres, iou_thres, classes, agnostic_nms, max_det=max_det)
+
+    # Process predictions
+    preds = []
+    for i, det in enumerate(pred):  # per image
+        
+        # im0 = im0.copy()
+        
+        if len(det):
+            # Rescale boxes from img_size to im0 size
+            det[:, :4] = scale_coords(im.shape[2:], det[:, :4], im0.shape).round()
+
+            for *xyxy, conf, cls in reversed(det):
+                face = im0[int(xyxy[1]):int(xyxy[3]),int(xyxy[0]):int(xyxy[2])]
+                face_img = cv2.cvtColor(face, cv2.COLOR_BGR2RGB)
+                im = test_transforms(face_img).unsqueeze_(0)
+                with torch.no_grad():
+                    y = age_model(im)
+                
+                age = y[0]
+
+                preds.append({"class": str(int(age)), "xmin": int(xyxy[0]), "ymin": int(xyxy[1]), "xmax": int(xyxy[2]),"ymax": int(xyxy[3]), "conf": float(conf)})
+                
+    return preds

yolov5/utils/datasets.py

@@ -87,6 +87,19 @@ def exif_transpose(image):
             image.info["exif"] = exif.tobytes()
     return image
 
+
+def pil_to_cv(pil_img, img_size=320, stride=32, auto=True):
+    np_img = np.array(pil_img)
+    
+    img0 = cv2.cvtColor(np_img, cv2.COLOR_RGB2BGR)
+
+    img = letterbox(img0, img_size, stride=stride, auto=auto)[0]
+    img = img.transpose((2, 0, 1))[::-1]  # HWC to CHW, BGR to RGB
+    img = np.ascontiguousarray(img)
+
+    return img, img0
+
+
 class LoadImages:
     # YOLOv5 image/video dataloader, i.e. `python detect.py --source image.jpg/vid.mp4`
     def __init__(self, path, img_size=640, stride=32, auto=True):