</> Code with Papers by Syntology

Datasets › HumanML3D

HumanML3D

Introduced by Chuan Guo et al. in Generating Diverse and Natural 3D Human Motions From Text1 Jan 2022 archive 2025-07-28

HumanML3D is a 3D human motion-language dataset that originates from a combination of HumanAct12 and Amass dataset. It covers a broad range of human actions such as daily activities (e.g., 'walking', 'jumping'), sports (e.g., 'swimming', 'playing golf'), acrobatics (e.g., 'cartwheel') and artistry (e.g., 'dancing'). Overall, HumanML3D dataset consists of 14,616 motions and 44,970 descriptions composed by 5,371 distinct words. The total length of motions amounts to 28.59 hours. The average motion length is 7.1 seconds, while average description length is 12 words.

HomepageBenchmarksPapers JSON

Benchmarks archive 2025-07-28

All 2 leaderboards whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.

First row (archive order)PaperCode
Motion Synthesis HumanML3D Motion Anything FID 0.028 Motion Anything: Any to Motion Generation steve-zeyu-zhang/MotionAnything 37 Compare
Motion Captioning HumanML3D ST-MLP BLEU-4 25.0 Guided Attention for Interpretable Motion Captioning rd20karim/m2t-interpretable 4 Compare

Papers archive 2025-07-28

30 shown of 33 papers with a leaderboard row on this dataset's benchmarks, newest first. The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset; the archive's count for this dataset is 201. The Syntology column is from Syntology's graph (read 2026-09-24), stated per sample; it is not part of any archive number.

DateSamples run Syntology
Motion Anything: Any to Motion Generation 1 1 10 Mar 2025 not harvested
MotionPCM: Real-Time Motion Synthesis with Phased Consistency Model 0 1 31 Jan 2025 not harvested
Free-T2M: Frequency Enhanced Text-to-Motion Diffusion Model With Consistency Loss 1 1 30 Jan 2025 not harvested
DisCoRD: Discrete Tokens to Continuous Motion via Rectified Flow Decoding 0 1 29 Nov 2024 not harvested
BiPO: Bidirectional Partial Occlusion Network for Text-to-Motion Synthesis 0 1 28 Nov 2024 not harvested
BAD: Bidirectional Auto-regressive Diffusion for Text-to-Motion Generation 1 2 17 Sep 2024 not harvested
MotionLCM: Real-time Controllable Motion Generation via Latent Consistency Model 1 1 30 Apr 2024 ran 0 of 1 samples (1 unverified; 1 pointer-only for licence)
MCM: Multi-condition Motion Synthesis Framework 1 1 19 Apr 2024 not harvested
BAMM: Bidirectional Autoregressive Motion Model 1 1 28 Mar 2024 not harvested
ParCo: Part-Coordinating Text-to-Motion Synthesis 1 1 27 Mar 2024 not harvested
Motion Mamba: Efficient and Long Sequence Motion Generation 1 1 12 Mar 2024 not harvested
GUESS:GradUally Enriching SyntheSis for Text-Driven Human Motion Generation 1 1 4 Jan 2024 not harvested
FineMoGen: Fine-Grained Spatio-Temporal Motion Generation and Editing 1 1 22 Dec 2023 ran 3 of 3 samples (0 unverified; 3 pointer-only for licence)
MMM: Generative Masked Motion Model 1 2 6 Dec 2023 ran 1 of 1 samples (0 unverified; 1 pointer-only for licence)
EMDM: Efficient Motion Diffusion Model for Fast and High-Quality Motion Generation 1 1 4 Dec 2023 ran 3 of 7 samples (4 unverified; 7 pointer-only for licence)
MoMask: Generative Masked Modeling of 3D Human Motions 1 1 29 Nov 2023 ran 8 of 11 samples (3 unverified)
Motion2Language, unsupervised learning of synchronized semantic motion segmentation 1 1 16 Oct 2023 not harvested
Guided Attention for Interpretable Motion Captioning 1 1 11 Oct 2023 not harvested
Act As You Wish: Fine-Grained Control of Motion Diffusion Model with Hierarchical Semantic Graphs 1 1 21 Sep 2023 not harvested
Fg-T2M: Fine-Grained Text-Driven Human Motion Generation via Diffusion Model 0 1 12 Sep 2023 not harvested
DiverseMotion: Towards Diverse Human Motion Generation via Discrete Diffusion 0 2 4 Sep 2023 not harvested
AttT2M: Text-Driven Human Motion Generation with Multi-Perspective Attention Mechanism 1 1 2 Sep 2023 ran 1 of 1 samples (0 unverified)
MotionGPT: Human Motion as a Foreign Language 3 1 26 Jun 2023 ran 1 of 9 samples (8 unverified)
Make-An-Animation: Large-Scale Text-conditional 3D Human Motion Generation 0 1 16 May 2023 not harvested
TM2D: Bimodality Driven 3D Dance Generation via Music-Text Integration 1 1 5 Apr 2023 ran 1 of 1 samples (0 unverified; 1 pointer-only for licence)
ReMoDiffuse: Retrieval-Augmented Motion Diffusion Model 1 1 3 Apr 2023 not harvested
T2M-GPT: Generating Human Motion from Textual Descriptions with Discrete Representations 1 3 15 Jan 2023 ran 5 of 7 samples (2 unverified)
Executing your Commands via Motion Diffusion in Latent Space 1 1 8 Dec 2022 ran 1 of 1 samples (0 unverified)
Diffusion Motion: Generate Text-Guided 3D Human Motion by Diffusion Model 0 1 22 Oct 2022 not harvested
Human Motion Diffusion Model 1 1 29 Sep 2022 ran 1 of 1 samples (0 unverified; 1 pointer-only for licence)

The full list of 33 is in the JSON twin.

Dataset loaders archive 2025-07-28

No loader listed in the archive.

Tasks archive 2025-07-28

Motion SynthesisMotion CaptioningMotion Generation

License archive 2025-07-28

No licence recorded in the archive. Absence here is not a statement about the dataset's terms.

Modalities archive 2025-07-28

Texts3D

Languages archive 2025-07-28

English

Variants archive 2025-07-28

  • HumanML3D

1 variant name, as the archive lists them.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections