Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386076369> ?p ?o ?g. }
- W4386076369 abstract "Our goal is to learn a video representation that is useful for downstream procedure understanding tasks in instructional videos. Due to the small amount of available annotations, a key challenge in procedure understanding is to be able to extract from unlabeled videos the procedural knowledge such as the identity of the task (e.g., ‘make latte’), its steps (e.g., ‘pour milk’), or the potential next steps given partial progress in its execution. Our main insight is that instructional videos depict sequences of steps that repeat between instances of the same or different tasks, and that this structure can be well represented by a Procedural Knowledge Graph (PKG), where nodes are discrete steps and edges connect steps that occur sequentially in the instructional activities. This graph can then be used to generate pseudo labels to train a video representation that encodes the procedural knowledge in a more accessible form to generalize to multiple procedure understanding tasks. We build a PKG by combining information from a text-based procedural knowledge database and an unlabeled instructional video corpus and then use it to generate training pseudo labels with four novel pre-training objectives. We call this PKG-based pre-training procedure and the resulting model Paprika, Procedure-Aware PRe-training for Instructional Knowledge Acquisition. We evaluate Paprika on COIN and CrossTask for procedure understanding tasks such as task recognition, step recognition, and step forecasting. Paprika yields a video representation that improves over the state of the art: up to 11.23% gains in accuracy in 12 evaluation settings. Implementation is available at https://github.com/salesforce/paprika." @default.
- W4386076369 created "2023-08-23" @default.
- W4386076369 creator A5018518655 @default.
- W4386076369 creator A5042646536 @default.
- W4386076369 creator A5065277392 @default.
- W4386076369 creator A5069589954 @default.
- W4386076369 creator A5073429776 @default.
- W4386076369 date "2023-06-01" @default.
- W4386076369 modified "2023-09-27" @default.
- W4386076369 title "Procedure-Aware Pretraining for Instructional Video Understanding" @default.
- W4386076369 cites W2422305492 @default.
- W4386076369 cites W2550462002 @default.
- W4386076369 cites W2798708692 @default.
- W4386076369 cites W2948242301 @default.
- W4386076369 cites W2952132648 @default.
- W4386076369 cites W2957775769 @default.
- W4386076369 cites W2960747818 @default.
- W4386076369 cites W2962795934 @default.
- W4386076369 cites W2963631366 @default.
- W4386076369 cites W2963814513 @default.
- W4386076369 cites W2964037671 @default.
- W4386076369 cites W2964094654 @default.
- W4386076369 cites W2980037812 @default.
- W4386076369 cites W2981851019 @default.
- W4386076369 cites W2984008963 @default.
- W4386076369 cites W2990720095 @default.
- W4386076369 cites W3011215845 @default.
- W4386076369 cites W3034215340 @default.
- W4386076369 cites W3035467150 @default.
- W4386076369 cites W3035635319 @default.
- W4386076369 cites W3103543556 @default.
- W4386076369 cites W3105232955 @default.
- W4386076369 cites W3105441977 @default.
- W4386076369 cites W3120784275 @default.
- W4386076369 cites W3145385912 @default.
- W4386076369 cites W3168640669 @default.
- W4386076369 cites W3174889475 @default.
- W4386076369 cites W3175300676 @default.
- W4386076369 cites W3176481196 @default.
- W4386076369 cites W3177173029 @default.
- W4386076369 cites W3199332348 @default.
- W4386076369 cites W3202074654 @default.
- W4386076369 cites W3203256294 @default.
- W4386076369 cites W3203711169 @default.
- W4386076369 cites W3207732590 @default.
- W4386076369 cites W4206418986 @default.
- W4386076369 cites W4210352519 @default.
- W4386076369 cites W4210871970 @default.
- W4386076369 cites W4213312439 @default.
- W4386076369 cites W4214507759 @default.
- W4386076369 cites W4214555767 @default.
- W4386076369 cites W4221142658 @default.
- W4386076369 cites W4221153256 @default.
- W4386076369 cites W4285105798 @default.
- W4386076369 cites W4312415723 @default.
- W4386076369 cites W4312463400 @default.
- W4386076369 cites W4312685541 @default.
- W4386076369 cites W4312864639 @default.
- W4386076369 cites W4312884055 @default.
- W4386076369 cites W4313053794 @default.
- W4386076369 cites W4313055276 @default.
- W4386076369 cites W4313118773 @default.
- W4386076369 cites W4313186260 @default.
- W4386076369 doi "https://doi.org/10.1109/cvpr52729.2023.01033" @default.
- W4386076369 hasPublicationYear "2023" @default.
- W4386076369 type Work @default.
- W4386076369 citedByCount "0" @default.
- W4386076369 crossrefType "proceedings-article" @default.
- W4386076369 hasAuthorship W4386076369A5018518655 @default.
- W4386076369 hasAuthorship W4386076369A5042646536 @default.
- W4386076369 hasAuthorship W4386076369A5065277392 @default.
- W4386076369 hasAuthorship W4386076369A5069589954 @default.
- W4386076369 hasAuthorship W4386076369A5073429776 @default.
- W4386076369 hasConcept C119857082 @default.
- W4386076369 hasConcept C132525143 @default.
- W4386076369 hasConcept C154945302 @default.
- W4386076369 hasConcept C162324750 @default.
- W4386076369 hasConcept C175154964 @default.
- W4386076369 hasConcept C17744445 @default.
- W4386076369 hasConcept C187736073 @default.
- W4386076369 hasConcept C199539241 @default.
- W4386076369 hasConcept C204321447 @default.
- W4386076369 hasConcept C23123220 @default.
- W4386076369 hasConcept C26517878 @default.
- W4386076369 hasConcept C2776359362 @default.
- W4386076369 hasConcept C2780451532 @default.
- W4386076369 hasConcept C2987255567 @default.
- W4386076369 hasConcept C38652104 @default.
- W4386076369 hasConcept C41008148 @default.
- W4386076369 hasConcept C49774154 @default.
- W4386076369 hasConcept C80444323 @default.
- W4386076369 hasConcept C94625758 @default.
- W4386076369 hasConceptScore W4386076369C119857082 @default.
- W4386076369 hasConceptScore W4386076369C132525143 @default.
- W4386076369 hasConceptScore W4386076369C154945302 @default.
- W4386076369 hasConceptScore W4386076369C162324750 @default.
- W4386076369 hasConceptScore W4386076369C175154964 @default.
- W4386076369 hasConceptScore W4386076369C17744445 @default.
- W4386076369 hasConceptScore W4386076369C187736073 @default.
- W4386076369 hasConceptScore W4386076369C199539241 @default.