Base inference

.gitattributes

@@ -25,3 +25,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zstandard filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+checkpoints_model_final_imagenet_40k_synthetic.pth filter=lfs diff=lfs merge=lfs -text

app.py

@@ -0,0 +1,34 @@
+import gradio as gr
+import torch
+import torchvision
+import logging
+from detectron2.engine import DefaultPredictor
+import cv2
+from detectron2.config import get_cfg
+from src.utils.visualizer import add_bboxes
+
+config_file="config.yaml"
+cfg = get_cfg()
+cfg.merge_from_file(config_file)
+cfg.MODEL.DEVICE="cpu"
+cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.5
+cfg.MODEL.WEIGHTS = "checkpoints_model_final_imagenet_40k_synthetic.pth.pth"
+
+def predict(
+    config_file, checkpoint_file, img_path
+):
+    predictor = DefaultPredictor(cfg)
+    im = cv2.imread(img_path)
+    output = predictor(im)
+    img = add_bboxes(im, output['instances'].pred_boxes, scores=output['instances'].scores)
+    return img
+
+title = "Pet Detection"
+description = "Demo for Indoor Pet Detection"
+examples = [['example.jpg']]
+
+
+gr.Interface(predict, inputs=gr.inputs.Image(type="file"), outputs=gr.outputs.Image(type="pil"),enable_queue=True, title=title,
+    description=description,
+    # article=article,
+    examples=examples).launch()

checkpoints_model_final_imagenet_40k_synthetic.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e71df17784bb168aec2655fe76c196ece86706467c18080d1dc72518ccb444f
+size 165717485

utils.py

@@ -0,0 +1,43 @@
+import cv2
+
+
+def draw_text(img, text,
+          font=cv2.FONT_HERSHEY_PLAIN,
+          x=0,
+          y=0,
+          font_scale=1,
+          font_thickness=2,
+          text_color=(0, 255, 0),
+          text_color_bg=(0, 0, 0)
+          ):
+
+    text_size, _ = cv2.getTextSize(text, font, font_scale, font_thickness)
+    text_w, text_h = text_size
+    cv2.rectangle(img, (x, y), (x + text_w, y + text_h), text_color_bg, -1)
+    cv2.putText(img, text, (x, y + text_h + font_scale - 1), font, font_scale, text_color, font_thickness)
+
+    return img
+
+
+def add_bboxes(img, bboxes, **kwargs):
+    label = "dog"
+    font_thickness=2
+    for idx, box in enumerate(bboxes):
+        x1, y1 = int(box[0]), int(box[1])
+        x2, y2 = int(box[2]), int(box[3])
+        if "scores" in kwargs.keys():
+            score = list(kwargs["scores"])[idx]
+            label = "{0:.2f}%".format(score*100)
+            font_thickness=1
+        img = cv2.rectangle(img, (x1, y1), (x2, y2), (36,255,12), 2)
+        img = draw_text(img, label, x=x1, y=y1-10, font_thickness=font_thickness)
+    return img
+
+
+def reformat_bbox(coord):
+    """
+    Converts XYWH to XYXY
+    """
+    x1, y1 = coord[0], coord[1]
+    x2, y2 = x1 + coord[2], y1 + coord[3]
+    return [x1, y1, x2, y2]