scaling demo output properly

edwin-pan · Mar 20, 2021 · d3ee090 · d3ee090
1 parent 195933c
commit d3ee090
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 1 deletion.
diff --git a/README.md b/README.md
@@ -76,6 +76,7 @@ python3 train.py --cfg <PATH-TO-CONFIG>
 
 ## Google Colab
 To quickly get a sense of how MetaHDR performs, we have provided a Google Colab notebook version of the demo that runs out of the box. The demo runs on the cloud for free.
+
 [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1WeyrYHh2BvRraUz9Xorr99Iy8CGM6_aM?usp=sharing)
 
 

diff --git a/demo.py b/demo.py
@@ -117,7 +117,7 @@ def main(args):
         HDR_reconst, test_ssim, test_psnr = evaluate_maml(meta_model, loss_func, training, testing, 0, cfg.EVAL.NUM_TASK_TR_ITER, device=device, model_type=cfg.TRAIN.MODEL, visualize_flag=True, visualize_dir=output_dir)
 
     # Save gamma corrected output image
-    io.imsave(f"{output_dir}/HDR{0:03d}.png", visualize_hdr_image(HDR_reconst).astype(np.uint8))
+    io.imsave(f"{output_dir}/HDR{0:03d}.png", (255*visualize_hdr_image(HDR_reconst)).astype(np.uint8))
 
     return
Original file line number	Diff line number	Diff line change
Expand Up		@@ -76,6 +76,7 @@ python3 train.py --cfg <PATH-TO-CONFIG>

		## Google Colab
		To quickly get a sense of how MetaHDR performs, we have provided a Google Colab notebook version of the demo that runs out of the box. The demo runs on the cloud for free.

		[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1WeyrYHh2BvRraUz9Xorr99Iy8CGM6_aM?usp=sharing)


Expand Down