From edd16c5d3f346a0ec009cec539fbcaf6d79b5111 Mon Sep 17 00:00:00 2001 From: Marc T Law Date: Mon, 23 Oct 2023 11:39:04 -0400 Subject: [PATCH] Update index.html --- docs/index.html | 395 +----------------------------------------------- 1 file changed, 5 insertions(+), 390 deletions(-) diff --git a/docs/index.html b/docs/index.html index 7e021c6..0898a34 100644 --- a/docs/index.html +++ b/docs/index.html @@ -1,390 +1,5 @@ - - - - - - - - - - - - - - Gated Shape CNN - - - - -
-
- Gated-SCNN -
- Gated Shape CNNs for Semantic Segmentation -
- -
- - - - - - - - - - - -
-
- Towaki Takikawa*1,2 -
- 		-
- David Acuna*1,3,4 -
- 		-
- Varun Jampani1 -
- 		-
- Sanja Fidler1,3,4 -
-
- -
- - - - - - - -
-
- 1NVIDIA -
- 		-
- 2University of Waterloo -
- 		-
- 3University of Toronto -
- 		-
- 4Vector Institute -
-
- - - - - -
-
- ICCV, 2019 -
-
-
- - - - - - -
-
- -
- 		-
-
-
-
-
- - -
-
- -
- - - - -
-

- - Current state-of-the-art methods for image segmentation form a dense image representation where the color, shape and texture information are all processed together inside a deep CNN. This however may not be ideal as they contain very different type of information relevant for recognition. We propose a new architecture that adds a shape stream to the classical CNN architecture. The two streams process the image in parallel, and their information gets fused in the very top layers. Key to this architecture is a new type of gates that connect the intermediate layers of the two streams. Specifically, we use the higher-level activations in the classical stream to gate the lower-level activations in the shape stream, effectively removing noise and helping the shape stream to only focus on processing the relevant boundary-related information. This enables us to use a very shallow architecture for the shape stream that operates on the image-level resolution. Our experiments show that this leads to a highly effective architecture that produces sharper predictions around object boundaries and significantly boosts performance on thinner and smaller objects. Our method achieves state-of-the-art performance on the Cityscapes benchmark, in terms of both mask (mIoU) and boundary (F-score) quality, improving by 2% and 4% over strong baselines. -

- - - - - - -
-
- -

News

- - - -
-
-
- -

Paper

- - - - -
Towaki Takikawa* , David Acuna* , Varun Jampani , Sanja Fidler
- (* denotes equal contribution)

- Gated-SCNN: Gated Shape CNNs for Semantic Segmentation

- - ICCV, 2019. (to appear)
-
-
- - - - - - - - - - - - - -
-
- [Preprint] -
- [Bibtex] -
- [Video] -
-
-
- -

GSCNN in a nutshell

- - -
- -
- -
- -
-
- -

Results


- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
-
-
-
-
-
- - Qualitative Segmentation Results - -
-
-
-
-
-
-
- - Qualitative Semantic Boundary Results - -
-
-
-
-
-
-
- - Quantitative Results - -
-
-
-
-
-
-
- - Evaluation at different distances, measured by crop factor. - -
-
-
-
- - - -
- - - + + +Redirecting to https://research.nvidia.com/labs/toronto-ai/GSCNN/ + +