Skip to content

Commit 6626965

Browse files
committed
fixed broken demos
1 parent c6760f2 commit 6626965

File tree

3 files changed

+4
-4
lines changed

3 files changed

+4
-4
lines changed

README.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -15,7 +15,7 @@ Zachary Teed and Jia Deng
1515
}
1616
```
1717

18-
**Initial Code Release:** This repo currently provides a single GPU implementation of our monocular SLAM system. It currently contains demos, training, and evaluation scripts.
18+
**Initial Code Release:** This repo currently provides a single GPU implementation of our monocular, stereo, and RGB-D SLAM systems. It currently contains demos, training, and evaluation scripts.
1919

2020

2121
## Requirements

demo.py

Lines changed: 3 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -82,6 +82,7 @@ def image_stream(imagedir, calib, stride):
8282
parser.add_argument("--backend_nms", type=int, default=3)
8383
args = parser.parse_args()
8484

85+
args.stereo = False
8586
torch.multiprocessing.set_start_method('spawn')
8687

8788
droid = None
@@ -92,10 +93,10 @@ def image_stream(imagedir, calib, stride):
9293
continue
9394

9495
if not args.disable_vis:
95-
show_image(image)
96+
show_image(image[0])
9697

9798
if droid is None:
98-
args.image_size = [image.shape[1], image.shape[2]]
99+
args.image_size = [image.shape[2], image.shape[3]]
99100
droid = Droid(args)
100101

101102
droid.track(t, image, intrinsics=intrinsics)

droid_slam/motion_filter.py

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -32,7 +32,6 @@ def __init__(self, net, video, thresh=2.5, device="cuda:0"):
3232
@torch.cuda.amp.autocast(enabled=True)
3333
def __context_encoder(self, image):
3434
""" context features """
35-
x = self.cnet(image)
3635
net, inp = self.cnet(image).split([128,128], dim=2)
3736
return net.tanh().squeeze(0), inp.relu().squeeze(0)
3837

0 commit comments

Comments
 (0)