bug fix

app.py

@@ -54,7 +54,7 @@ def inference(class_names, proposal_gen, granularity, input_img):
 
 
 examples = [['Saturn V, toys, desk, sunflowers, white roses, chrysanthemums, carnations, green dianthus', 'Segment_Anything', 0.8, './resources/demo_samples/sample_01.jpeg'],
-            ['red bench, yellow bench, black bench, blue bench, brown bench, green bench, red chair, blue chair, yellow chair, green chair', 'Segment_Anything', 0.8, './resources/demo_samples/sample_04.png'],
+            ['red bench, yellow bench, blue bench, brown bench, green bench, blue chair, yellow chair, green chair', 'Segment_Anything', 0.8, './resources/demo_samples/sample_04.png'],
             ['Saturn V, toys, blossom', 'MaskFormer', 1.0, './resources/demo_samples/sample_01.jpeg'],
             ['Oculus, Ukulele', 'MaskFormer', 1.0, './resources/demo_samples/sample_03.jpeg'],
             ['Golden gate, yacht', 'MaskFormer', 1.0, './resources/demo_samples/sample_02.jpeg'],]

open_vocab_seg/utils/predictor.py

@@ -150,7 +150,7 @@ class SAMVisualizationDemo(object):
 
         self.parallel = parallel
         self.granularity = granularity
-        sam = sam_model_registry["vit_h"](checkpoint=sam_path)
+        sam = sam_model_registry["vit_h"](checkpoint=sam_path).cuda()
         self.predictor = SamAutomaticMaskGenerator(sam)
         self.clip_model, _, _ = open_clip.create_model_and_transforms('ViT-L-14', pretrained=ovsegclip_path)
         self.clip_model.cuda()
@@ -190,8 +190,8 @@ class SAMVisualizationDemo(object):
         text = open_clip.tokenize(txts)
 
         with torch.no_grad(), torch.cuda.amp.autocast():
-            image_features = self.clip_model.encode_image(imgs)
-            text_features = self.clip_model.encode_text(text)
+            image_features = self.clip_model.encode_image(imgs.half())
+            text_features = self.clip_model.encode_text(text.half())
             image_features /= image_features.norm(dim=-1, keepdim=True)
             text_features /= text_features.norm(dim=-1, keepdim=True)