babyview-dino-vits16 — DINOv3 ViT-S trained from scratch on children's egocentric video

A 22M-parameter DINOv3 ViT-S/16 trained from scratch, self-supervised, on the BabyView corpus (Long et al., 2025): ~9.7M frames sampled at 1 fps from 2,701 hours of head-mounted-camera video of 51 children (ages 3–54 months) — release 2026.1. No internet images, no supervision, no pretrained initialization: this model has only ever seen a child's visual world.

Trained as part of "Scaling cross-situational word learning on children's naturalistic experiences" (Frank lab, Stanford; paper in prep) to test whether developmental-scale visual experience supports the representations needed for word learning.

Training

Stock DINOv3 recipe (github.com/awwkl/dinov3 fork of facebookresearch/dinov3), from scratch: global batch 512 (no gradient accumulation), 200k iterations (~102M samples seen), lr rule sqrt_wrt_1024, bf16, RoPE, Sinkhorn-Knopp centering + KoLeo + iBOT. Released weights are the EMA teacher backbone at iteration 199,999.

Evaluation (from the paper's probes)

• Konkle object prototype 4AFC (head-free object separability): high 90s — near internet- pretrained models.
• Cross-situational word-learning 4AFC after frozen-readout training on the full BabyView corpus: 39.6 (internet-pretrained DINOv3-S: high 70s–low 80s; chance 25). The persistent gap to internet pretraining — despite object separability — is the paper's central result about visual experience and word learning.

Two artifacts in this repo

1. HF-transformers weights (config.json + model.safetensors): load with AutoModel.from_pretrained. Note: the training run stored RoPE periods in bf16; the HF port recomputes them in fp32, so features differ microscopically from the training artifact (word-probe 29.1 ± 2.7 vs native 31.0 on our 100k probe — statistically indistinguishable).
2. vits_teacher_backbone_native.pt: the exact training artifact (native dinov3 state dict). Load with the dinov3 repo: build_model(cfg.student, only_teacher=True) then load_state_dict. The paper's numbers use this artifact.

Access

These weights are gated: access is limited to researchers who have been granted access to the BabyView dataset, because BabyView contains egocentric video of children in home and school environments.

1. Request access to BabyView through Databrary: https://www.databrary.org/volume/1882
2. Once approved, click Request access on this page, then email mcfrank@stanford.edu with your Hugging Face username and a screenshot showing your BabyView access on Databrary.
3. We will approve your request.

More about the dataset: https://babyview-project.github.io/dataset/

Data statement

Model weights only. BabyView videos/frames are restricted human-subjects data and are NOT distributed here; corpus access is governed by the BabyView data-sharing process (Long et al., 2025, arXiv:2406.10447).

Acknowledgments

Training recipe and codebase groundwork by Khai Loong Aw (awwkl); BabyView corpus by the Frank lab and the BabyView team.