SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2103.11264 · CVPR · 2021

Temporally-Weighted Hierarchical Clustering for Unsupervised Action Segmentation

Luc Van Gool, K Leuven, M Saquib Sarfraz, Naila Murray, Rainer Stiefelhagen, Vivek Sharma, Ali Diba, Daimler Tss

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 7 functions out of this paper's own repositories and ran 5 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
ssarfraz/FINCH-Clustering — 4 of 6
copy not recorded — 1 of 1
FunctionStatusWhere it lives
FINCH Ran ssarfraz/FINCH-Clustering/TW-FINCH/python/twfinch.py
code served (permissive licence) · get_code("f333f36a54cff274")
clust_rank Ran ssarfraz/FINCH-Clustering/TW-FINCH/python/twfinch.py
code served (permissive licence) · get_code("7fc40801f2b43dfc")
cool_mean Ran ssarfraz/FINCH-Clustering/TW-FINCH/python/twfinch.py
code served (permissive licence) · get_code("8b82e4cf700e1ffd")
get_clust Ran this paper's copy was not recorded; identical code first harvested from ssarfraz/FINCH-CLustering
pointer only · get_code("a13645c6a637da8d")
req_numclust Ran ssarfraz/FINCH-Clustering/TW-FINCH/python/twfinch.py
code served (permissive licence) · get_code("8658952e06494ab4")
get_merge Not yet run ssarfraz/FINCH-Clustering/TW-FINCH/python/twfinch.py
code served (permissive licence) · get_code("0aca21662917fe20")
update_adj Not yet run ssarfraz/FINCH-Clustering/TW-FINCH/python/twfinch.py
code served (permissive licence) · get_code("34c60cf7ca7ee21e")

Repositories linked to this paper

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

Abstract

Action segmentation refers to inferring boundaries of semantically consistent visual concepts in videos and is an important requirement for many video understanding tasks. For this and other video understanding tasks, supervised approaches have achieved encouraging performance but require a high volume of detailed frame-level annotations. We present a fully automatic and unsupervised approach for segmenting actions in a video that does not require any training. Our proposal is an effective temporally-weighted hierarchical clustering algorithm that can group semantically consistent frames of the video. Our main finding is that representing a video with a 1-nearest neighbor graph by taking into account the time progression is sufficient to form semantically and temporally consistent clusters of frames where each cluster may represent some action in the video. Additionally, we establish strong unsupervised baselines for action segmentation and show significant performance improvements over published unsupervised methods on five challenging action segmentation datasets. Our code is available. 1

For agents

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

get_harvested_code_for_paper("2103.11264")
get_code_for_paper("2103.11264")
have("2103.11264")

Connect an agent — have() is free.