modify for PIL

app.py

@@ -37,26 +37,27 @@ ckpt_filenmae = "groundingdino_swint_ogc.pth"
 
 
 def detection(image):
-    sub_images = processPanorama(image)
-    processed_images = [np.array(sub_image) for sub_image in sub_images]
+    sub_images = process_panorama(image)
+    predict_images = []
+    for sub_image in sub_images:
+        predict_images.append(run_grounding(sub_image))
 
-    return processed_images
+    return predict_images
 
 
-def processPanorama(image):
+def process_panorama(image):
     equ = E2P.Equirectangular(image)
-    FOV = picture_fov
     y_axis = 0
 
     sub_images = []
     while y_axis <= 0:
         z_axis = -150
         while z_axis <= 90:
-            img = equ.GetPerspective(FOV, z_axis, y_axis, picture_height, picture_width)
+            img = equ.GetPerspective(picture_fov, z_axis, y_axis, picture_height, picture_width)
             # cv2.imwrite(f'{directory_name}_{z_axis}z.jpg', img)
             sub_images.append(img)
-            z_axis += FOV
-        y_axis += FOV
+            z_axis += picture_fov
+        y_axis += picture_fov
     return sub_images
 
 
@@ -94,9 +95,14 @@ def image_transform_grounding_for_vis(init_image):
 model = load_model_hf(config_file, ckpt_repo_id, ckpt_filenmae)
 
 
-def run_grounding(input_image, grounding_caption, box_threshold, text_threshold):
-    init_image = input_image.convert("RGB")
+def run_grounding(input_image):
+    cv2_img_rgb = cv2.cvtColor(input_image, cv2.COLOR_BGR2RGB)
+    pil_img = Image.fromarray(cv2_img_rgb)
+    init_image = pil_img.convert("RGB")
     original_size = init_image.size
+    grounding_caption = "traffic sign"
+    box_threshold = 0.25
+    text_threshold = 0.25
 
     _, image_tensor = image_transform_grounding(init_image)
     image_pil: Image = image_transform_grounding_for_vis(init_image)
@@ -123,7 +129,7 @@ if __name__ == "__main__":
 
             with gr.Column():
                 gallery = gr.Gallery(label="Detection Results").style(
-                    columns=[3], preview=False, object_fit="none")
+                    columns=[3], preview=True, object_fit="none")
 
         run_button.click(fn=detection, inputs=[
             input_image], outputs=[gallery])