SOTAVerified

Motion Captioning

Generating textual description for human motion.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–10 of 11 papers

TitleStatusHype
GigaHands: A Massive Annotated Dataset of Bimanual Hand Activities—0
MoTe: Learning Motion-Text Diffusion Model for Multiple Generation Tasks—0
MotionGPT-2: A General-Purpose Motion-Language Model for Motion Generation and Understanding—0
LaMP: Language-Motion Pretraining for Motion Generation, Retrieval, and Captioning—0
Transformer with Controlled Attention for Synchronous Motion CaptioningCode0
Motion-Agent: A Conversational Framework for Human Motion Generation with LLMsCode2
Digital Life Project: Autonomous 3D Characters with Social Intelligence—0
Motion2Language, unsupervised learning of synchronized semantic motion segmentationCode1
Guided Attention for Interpretable Motion CaptioningCode0
MotionGPT: Human Motion as a Foreign LanguageCode3
Show:102550
← PrevPage 1 of 2Next →
All datasetsHumanML3DKIT Motion-Language

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1ST-MLPBLEU-425—Unverified
2MLP+GRUBLEU-423.4—Unverified
3TM2TBLEU-422.3—Unverified
4MotionGPTBLEU-412.47—Unverified
#ModelMetricClaimedVerifiedStatus
1MLP+GRUBLEU-425.4—Unverified
2ST-MLPBLEU-424.4—Unverified
3TM2TBLEU-418.4—Unverified