Matches in SemOpenAlex for { <https://semopenalex.org/work/W4225517595> ?p ?o ?g. }
- W4225517595 endingPage "3094" @default.
- W4225517595 startingPage "3081" @default.
- W4225517595 abstract "Humans have the inherent advantage of understanding action intention, while it is an enormous challenge to train the machine to localize unintentional action in videos due to the lack of reliable annotations for stable training. The annotations of unintentional action are unreliable since different annotators are affected by their subjective appraisals and intrinsic ambiguity, which brings heavy difficulties for the training. To address this issue, we propose a probabilistic framework for unintentional action localization by modeling the uncertainty of annotations. Our framework consists of two main components, including Temporal Label Aggregation (TLA) and Dense Probabilistic Localization (DPL). We first formulate each annotated failure moment as a temporal label distribution. Then we propose a TLA component to aggregate temporal label distributions of different failure moments in an online manner and generate dense probabilistic supervision. Based on TLA, We further develop a DPL component to jointly train three heads (i.e., probabilistic dense classification, probabilistic temporal detection, and probabilistic regression) with different supervision granularities and make them highly collaborative. We evaluate our approach on the largest unintentional action dataset OOPS and demonstrate that our approach can achieve significant improvement over the baseline and state-of-the-art methods." @default.
- W4225517595 created "2022-05-05" @default.
- W4225517595 creator A5069747026 @default.
- W4225517595 creator A5082393074 @default.
- W4225517595 creator A5082605318 @default.
- W4225517595 creator A5088257057 @default.
- W4225517595 creator A5090079801 @default.
- W4225517595 date "2022-01-01" @default.
- W4225517595 modified "2023-10-16" @default.
- W4225517595 title "Probabilistic Temporal Modeling for Unintentional Action Localization" @default.
- W4225517595 cites W1993720331 @default.
- W4225517595 cites W2022566595 @default.
- W4225517595 cites W2066454034 @default.
- W4225517595 cites W2118370524 @default.
- W4225517595 cites W2169662424 @default.
- W4225517595 cites W2183341477 @default.
- W4225517595 cites W2341058432 @default.
- W4225517595 cites W2342662179 @default.
- W4225517595 cites W2399941526 @default.
- W4225517595 cites W2507009361 @default.
- W4225517595 cites W2553156677 @default.
- W4225517595 cites W2579021424 @default.
- W4225517595 cites W2587789887 @default.
- W4225517595 cites W2593722617 @default.
- W4225517595 cites W2607566495 @default.
- W4225517595 cites W2660006607 @default.
- W4225517595 cites W2755876276 @default.
- W4225517595 cites W2777342313 @default.
- W4225517595 cites W2884002012 @default.
- W4225517595 cites W2905246249 @default.
- W4225517595 cites W2950534130 @default.
- W4225517595 cites W2952435096 @default.
- W4225517595 cites W2952506564 @default.
- W4225517595 cites W2962677524 @default.
- W4225517595 cites W2962876901 @default.
- W4225517595 cites W2962934715 @default.
- W4225517595 cites W2963091558 @default.
- W4225517595 cites W2963240734 @default.
- W4225517595 cites W2963247196 @default.
- W4225517595 cites W2963321993 @default.
- W4225517595 cites W2963524571 @default.
- W4225517595 cites W2963610939 @default.
- W4225517595 cites W2964008341 @default.
- W4225517595 cites W2964214371 @default.
- W4225517595 cites W2964216549 @default.
- W4225517595 cites W2964248614 @default.
- W4225517595 cites W2974602280 @default.
- W4225517595 cites W2983918066 @default.
- W4225517595 cites W2990503944 @default.
- W4225517595 cites W2990759760 @default.
- W4225517595 cites W3034504038 @default.
- W4225517595 cites W3034623254 @default.
- W4225517595 cites W3034630387 @default.
- W4225517595 cites W3035714233 @default.
- W4225517595 cites W3069380482 @default.
- W4225517595 cites W3096840866 @default.
- W4225517595 cites W3100481960 @default.
- W4225517595 cites W3129684859 @default.
- W4225517595 cites W3153639479 @default.
- W4225517595 cites W3157181916 @default.
- W4225517595 cites W3166166117 @default.
- W4225517595 cites W3168600998 @default.
- W4225517595 cites W3168984673 @default.
- W4225517595 cites W3169077988 @default.
- W4225517595 cites W3170074351 @default.
- W4225517595 cites W3172990466 @default.
- W4225517595 cites W3173229911 @default.
- W4225517595 cites W3173345775 @default.
- W4225517595 cites W3174226175 @default.
- W4225517595 cites W3175716777 @default.
- W4225517595 cites W3176212189 @default.
- W4225517595 cites W3176640650 @default.
- W4225517595 cites W3177339245 @default.
- W4225517595 cites W3177433885 @default.
- W4225517595 cites W3178036208 @default.
- W4225517595 cites W3189800722 @default.
- W4225517595 cites W4285453420 @default.
- W4225517595 doi "https://doi.org/10.1109/tip.2022.3163544" @default.
- W4225517595 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/35389866" @default.
- W4225517595 hasPublicationYear "2022" @default.
- W4225517595 type Work @default.
- W4225517595 citedByCount "4" @default.
- W4225517595 countsByYear W42255175952022 @default.
- W4225517595 countsByYear W42255175952023 @default.
- W4225517595 crossrefType "journal-article" @default.
- W4225517595 hasAuthorship W4225517595A5069747026 @default.
- W4225517595 hasAuthorship W4225517595A5082393074 @default.
- W4225517595 hasAuthorship W4225517595A5082605318 @default.
- W4225517595 hasAuthorship W4225517595A5088257057 @default.
- W4225517595 hasAuthorship W4225517595A5090079801 @default.
- W4225517595 hasConcept C114289077 @default.
- W4225517595 hasConcept C119857082 @default.
- W4225517595 hasConcept C121332964 @default.
- W4225517595 hasConcept C124101348 @default.
- W4225517595 hasConcept C154945302 @default.
- W4225517595 hasConcept C159985019 @default.
- W4225517595 hasConcept C168167062 @default.
- W4225517595 hasConcept C192562407 @default.