Action Anticipation

Next action anticipation is defined as observing 1, ... , T frames and predicting the action that happens after a gap of T_a seconds. It is important to note that a new action starts after T_a seconds that is not seen in the observed frames. Here T_a=1 second.

Papers

Recently Added Most Hyped Most Active Needs Verification Most Verified

Showing 1–10 of 110 papers

Title	Date	Tasks	Status	Hype
V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning	Jun 11, 2025	Action AnticipationLarge Language Model	CodeCode Available	7
EgoVideo: Exploring Egocentric Foundation Model and Downstream Adaptation	Jun 26, 2024	Action AnticipationAction Recognition	CodeCode Available	2
EgoExoLearn: A Dataset for Bridging Asynchronous Ego- and Exo-centric View of Procedural Activities in Real World	Mar 24, 2024	Action AnticipationAction Quality Assessment	CodeCode Available	2
Future Transformer for Long-term Action Anticipation	May 27, 2022	Action AnticipationLong Term Action Anticipation	CodeCode Available	1
Anticipative Feature Fusion Transformer for Multi-Modal Action Anticipation	Oct 23, 2022	Action Anticipation	CodeCode Available	1
Action Scene Graphs for Long-Form Understanding of Egocentric Videos	Dec 6, 2023	Action AnticipationForm	CodeCode Available	1
A Dynamic Spatial-temporal Attention Network for Early Anticipation of Traffic Accidents	Jun 18, 2021	Accident AnticipationAction Anticipation	CodeCode Available	1
Action Anticipation with Goal Consistency	Jun 26, 2023	Action Anticipation	CodeCode Available	1
AntGPT: Can Large Language Models Help Long-term Action Anticipation from Videos?	Jul 31, 2023	Action Anticipationcounterfactual	CodeCode Available	1
Anticipative Video Transformer	Jun 3, 2021	Action Anticipation	CodeCode Available	1

Show:10 25 50

← PrevPage 1 of 11Next →

All datasets EPIC-KITCHENS-100 EPIC-KITCHENS-100 (test)EPIC-KITCHENS-55 (Seen test set (S1))EPIC-KITCHENS-55 (Unseen test set (S2)EGTEA Assembly101 EgoExoLearn 50 Salads

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	PlausiVL	Recall@5	27.6	—	Unverified
2	InAViT	Recall@5	25.89	—	Unverified
3	UADT	Recall@5	23	—	Unverified
4	S-GEAR	Recall@5	19.9	—	Unverified
5	AFFT	Recall@5	18.5	—	Unverified
6	MeMViT-24	Recall@5	17.7	—	Unverified
7	AVT+	Recall@5	15.9	—	Unverified
8	TempAgg	Recall@5	14.73	—	Unverified
9	RU-LSTM	Recall@5	13.94	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	InAViT	recall@5	23.75	—	Unverified
2	AVT++	recall@5	16.7	—	Unverified
3	AFFT	recall@5	14.9	—	Unverified
4	Abstract Goal	recall@5	14.29	—	Unverified
5	AVT+	recall@5	12.6	—	Unverified
6	TempAgg	recall@5	12.6	—	Unverified
7	RULSTM	recall@5	11.2	—	Unverified
8	TBN	recall@5	11	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Abstract Goal	Top 1 Accuracy - Act.	22.03	—	Unverified
2	AVT+	Top 1 Accuracy - Act.	16.84	—	Unverified
3	ImagineRNN	Top 1 Accuracy - Act.	14.66	—	Unverified
4	RULSTM [24, 23]	Top 1 Accuracy - Act.	14.39	—	Unverified
5	ED	Top 1 Accuracy - Act.	8.08	—	Unverified
6	ATSN	Top 1 Accuracy - Act.	6	—	Unverified
7	2SCNN	Top 1 Accuracy - Act.	4.32	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Abstract Goal	Top 1 Accuracy - Act.	13.28	—	Unverified
2	AVT+	Top 1 Accuracy - Act.	10.41	—	Unverified
3	ImagineRNN	Top 1 Accuracy - Act.	9.25	—	Unverified
4	RULSTM [24, 23]	Top 1 Accuracy - Act.	8.16	—	Unverified
5	ED	Top 1 Accuracy - Act.	2.65	—	Unverified
6	ATSN	Top 1 Accuracy - Act.	2.39	—	Unverified
7	2SCNN	Top 1 Accuracy - Act.	2.29	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	UADT	Top-1 Accuracy	68.4	—	Unverified
2	InAViT	Top-1 Accuracy	67.8	—	Unverified
3	Abstract Goal	Top-1 Accuracy	49.8	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Goal Consistency	Verbs Recall@5	60.04	—	Unverified
2	TempAgg	Verbs Recall@5	59.11	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Action anticipation baseline (co-training, with gaze)	Accuracy	45.45	—	Unverified
2	Action anticipation baseline (co-training, no gaze)	Accuracy	38.7	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	UADT	Top-1 Accuracy	62.7	—	Unverified