SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2005.02551 · 2020

CascadePSP: Toward Class-Agnostic and Very High-Resolution Segmentation via Global and Local Refinement

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 7 functions out of this paper's own repositories and ran 2 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
hkchengrex/CascadePSP canonical 2 of 7
FunctionStatusWhere it lives
color_map Ran hkchengrex/CascadePSP/eval_post_ade.py
code served (permissive licence) · get_code("fcbd501f30a007d4")
conv3x3 Ran hkchengrex/CascadePSP/models/psp/extractors.py
code served (permissive licence) · get_code("48f5a5ec1d5dd2ef")
get_bb_position Not yet run hkchengrex/CascadePSP/dataset/make_bb_trans.py
code served (permissive licence) · get_code("58be3f0953cafaef")
get_iu Not yet run hkchengrex/CascadePSP/eval_post.py
code served (permissive licence) · get_code("f5f76587ff3a7654")
is_bb_overlap Not yet run hkchengrex/CascadePSP/dataset/make_bb_trans.py
code served (permissive licence) · get_code("5894805c2d7c5b80")
safe_forward Not yet run hkchengrex/CascadePSP/eval_helper.py
code served (permissive licence) · get_code("9233f30dee14c56f")
scale_bb_by Not yet run hkchengrex/CascadePSP/dataset/make_bb_trans.py
code served (permissive licence) · get_code("a190282758f268f5")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

State-of-the-art semantic segmentation methods were almost exclusively trained on images within a fixed resolution range. These segmentations are inaccurate for very high-resolution images since using bicubic upsampling of low-resolution segmentation does not adequately capture high-resolution details along object boundaries. In this paper, we propose a novel approach to address the high-resolution segmentation problem without using any high-resolution training data. The key insight is our CascadePSP network which refines and corrects local boundaries whenever possible. Although our network is trained with low-resolution segmentation data, our method is applicable to any resolution even for very high-resolution images larger than 4K. We present quantitative and qualitative studies on different datasets to show that CascadePSP can reveal pixel-accurate segmentation boundaries using our novel refinement module without any finetuning. Thus, our method can be regarded as class-agnostic. Finally, we demonstrate the application of our model to scene parsing in multi-class segmentation.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2005.02551")
get_code_for_paper("2005.02551")
have("2005.02551")

Connect an agent — have() is free.