Every9D-21M supplies 21.8M real-world 9D pose annotations for 700 everyday categories by propagating manual canonical poses through cross-instance alignment in object-centric videos and verifying them multiview.
super hub Mixed citations
Title resolution pending
Mixed citation behavior. Most common role is background (43%).
597
Pith papers citing it
4
external citations · Pith
Background
43% of classified citations
abstract
Self-supervised learning holds the promise of eliminating the need for manual data annotation, enabling models to scale effortlessly to massive datasets and larger architectures. By not being tailored to specific tasks or domains, this training paradigm has the potential to learn visual representations from diverse sources, ranging from natural to aerial images -- using a single algorithm. This technical report introduces DINOv3, a major milestone toward realizing this vision by leveraging simple yet effective strategies. First, we leverage the benefit of scaling both dataset and model size by careful data preparation, design, and optimization. Second, we introduce a new method called Gram anchoring, which effectively addresses the known yet unsolved issue of dense feature maps degrading during long training schedules. Finally, we apply post-hoc strategies that further enhance our models' flexibility with respect to resolution, model size, and alignment with text. As a result, we present a versatile vision foundation model that outperforms the specialized state of the art across a broad range of settings, without fine-tuning. DINOv3 produces high-quality dense features that achieve outstanding performance on various vision tasks, significantly surpassing previous self- and weakly-supervised foundation models. We also share the DINOv3 suite of vision models, designed to advance the state of the art on a wide spectrum of tasks and data by providing scalable solutions for diverse resource constraints and deployment scenarios.
hub tools
citation-role summary
method 38
background 37
baseline 9
dataset 4
other 1
citation-polarity summary
claims ledger
- abstract Self-supervised learning holds the promise of eliminating the need for manual data annotation, enabling models to scale effortlessly to massive datasets and larger architectures. By not being tailored to specific tasks or domains, this training paradigm has the potential to learn visual representations from diverse sources, ranging from natural to aerial images -- using a single algorithm. This technical report introduces DINOv3, a major milestone toward realizing this vision by leveraging simple yet effective strategies. First, we leverage the benefit of scaling both dataset and model size by
authors
co-cited works
representative citing papers
LeJEPA achieves linear identifiability of latent variables uniquely when the latents are Gaussian in worlds with stationary additive-noise transitions.
Dimension d = O(m^{-2} log n) nearly achieves the optimal margin m^rd(+∞, A) for retrieval embeddings, with matching lower bounds showing d = O(k log(n/k)) suffices and is necessary for m = Θ(k^{-1/2}) on k-sparse query matrices.
RICA replaces ICA's global generative model with local Riemannian geometry, introducing a disentanglement tensor based on the Hessian of the log-likelihood and Ricci curvature to measure pointwise disentanglement, which recovers sources across manifolds in controlled tests.
CanViT is the first task- and policy-agnostic AVFM pretrained via passive-to-active dense latent distillation on 13.2M scenes and 1B random glimpses, achieving 38.5% ADE20K mIoU in one glimpse and 84.5% ImageNet-1k top-1 after fine-tuning.
ConceptPose delivers state-of-the-art zero-shot relative pose estimation by matching open-vocabulary 3D concept vectors derived from VLM saliency maps, beating the strongest baseline by 62% in ADD(-S) without training.
A 5B-parameter latent diffusion model generates real-time four-player Rocket League matches conditioned on all players' actions, staying stable far beyond its training horizon.
A training-free prototype memory-guided framework for multi-class prenatal ultrasound anomaly classification and localization using few reference images per class, validated on a 9-category multi-center dataset.
Introduces the CHC dataset of 3 m continuous canopy height differences with uncertainties and the uncertainty-aware change regression task for fine-tuning GFMs on PlanetScope time series.
Active Spatial Guidance replaces injected positional embeddings in ViTs with a training-only 2D coordinate regression loss on final-layer tokens, yielding better results than learned absolute or rotary embeddings on ImageNet-100, ADE20K, and Hypersim under matched training.
GEAR jointly trains VQ tokenizer and AR generator end-to-end via dual hard/soft read-out and representation alignment, achieving up to 10x faster ImageNet gFID convergence than LlamaGen-REPA while generalizing across quantizers and to text-to-image.
SemCityLoc achieves aerial 6DoF localization via semantic-geometric alignment of monocular depth and surfaces with LoD1-LoD3 city models, cutting mean positional error to 2.62m and boosting recall up to 36% on a new real-world benchmark.
OctoSense supplies a large multimodal robotics dataset and a late-fusion masked autoencoder that runs fast and outperforms image-only models on optical flow, depth, segmentation, and ego-motion tasks while remaining robust under sensor degradation.
Introduces SmellNet-V synthetic visuo-olfactory dataset and See & Sniff self-supervised framework that learns aligned representations and produces smell saliency maps.
Block-sparse featurizers recover visual concepts as two- to four-dimensional manifolds and describe activations more compactly than direction-based methods via minimum-description-length comparison.
MATCH is the first flow matching method for multi-view anomaly detection, reporting SOTA results on Real-IAD and the first comprehensive evaluation on MANTA-Tiny while enabling real-time use by omitting the divergence term.
REDI-Match uses rotation-equivariant distillation to transfer VFM semantics into a strictly equivariant encoder plus an entropy-driven alignment module, claiming SOTA accuracy and 1.9x speed on rotation-heavy benchmarks.
HeRA aligns least-aligned attention heads in MLLMs using an MKNN-based contrastive objective to preserve cross-modal topological structure, yielding gains on vision-centric tasks and reduced hallucinations across 18 benchmarks.
UniverSat is a ViT-style model with a universal patch encoder enabling self-supervised training on heterogeneous multimodal Earth observation data from varying resolutions and sensors.
SSL representations for echo LV segmentation and EF estimation recover substantially more task information with expressive downstream probes than with heuristic extraction, nearing supervised U-Net performance on EchoNet-Dynamic.
HiMatch-AD proposes DINOv3-driven hierarchical matching with uncertainty-based fusion for training-free medical anomaly detection and reports outperformance on the BMAD benchmark.
A method that treats 3D box pairs as exact transformation specs, adds a depth-aware floor reference, and trains an image generator on synthetic scenes plus Objectron videos to perform large 3D edits on real photographs.
FlowBender introduces closed-loop training that lets conditional flow models learn correction policies from their own task-specific alignment errors, outperforming supervised and guidance baselines on fidelity and plausibility.
EFIQA uses unsupervised masked anatomical inpainting to learn normal fundus structures and produces spatial quality maps via a shallow adapter on a frozen foundation model without any quality supervision.
citing papers explorer
-
Is Dimensionality a Barrier for Retrieval Models?
Dimension d = O(m^{-2} log n) nearly achieves the optimal margin m^rd(+∞, A) for retrieval embeddings, with matching lower bounds showing d = O(k log(n/k)) suffices and is necessary for m = Θ(k^{-1/2}) on k-sparse query matrices.
-
H2G: Hierarchy-Aware Hyperbolic Grouping for 3D Scenes
H2G distills 2D foundation-model affinities into a Lorentz hyperbolic feature field that represents hierarchical 3D groupings at multiple granularities.
-
Adaptive Context Matters: Towards Provable Multi-Modality Guidance for Super-Resolution
The paper provides the first theoretical analysis of multi-modal super-resolution and proposes M³ESR, a mixture-of-experts framework with spatially dynamic and temporally adaptive modality weighting that improves generalization and semantic consistency.
-
AdaptSplat: Adapting Vision Foundation Models for Feed-Forward 3D Gaussian Splatting
AdaptSplat adds a Frequency-Preserving Adapter to vision foundation models to boost high-frequency fidelity and cross-domain performance in feed-forward 3D Gaussian Splatting.
-
Differentiable Ray Tracing with Gaussians for Unified Radio Propagation Simulation and View Synthesis
Embedding Gaussian primitives into a ray tracing structure enables unified radio propagation simulation and view synthesis from visual-only reconstructions.
-
Velocity-Space 3D Asset Editing
VS3D performs local 3D asset editing by injecting reconstruction-anchored source signals, partial-mean guidance, and twin-agreement residuals into the velocity sampler to control edit strength and preserve identity.
-
OA-WAM: Object-Addressable World Action Model for Robust Robot Manipulation
OA-WAM uses persistent address vectors and dynamic content vectors in object slots to enable addressable world-action prediction, improving robustness on manipulation benchmarks under scene changes.
-
Divide and Conquer: Decoupled Representation Alignment for Multimodal World Models
M²-REPA decouples modality-specific features from diffusion intermediates and aligns them to complementary expert foundation models via a multi-modal alignment loss and modality-specific decoupling regularization for improved multimodal video generation.
-
Large-scale Codec Avatars: The Unreasonable Effectiveness of Large-scale Avatar Pretraining
Pretraining on 1M wild videos followed by post-training on curated data yields high-fidelity feedforward 3D avatars that generalize across identities, clothing, and lighting with emergent relightability and loose-garment support.
-
TripVVT: A Large-Scale Triplet Dataset and a Coarse-Mask Baseline for In-the-Wild Video Virtual Try-On
A new large-scale triplet dataset and diffusion transformer model using coarse human masks deliver improved video virtual try-on quality and generalization in challenging real-world conditions.
-
UniFixer: A Universal Reference-Guided Fixer for Diffusion-Based View Synthesis
UniFixer is a universal reference-guided framework that fixes spatial, temporal, and backbone-related degradations in diffusion-based view synthesis via coarse-to-fine modules and achieves zero-shot SOTA results on novel view synthesis and stereo conversion.
-
MULTI: Disentangling Camera Lens, Sensor, View, and Domain for Novel Image Generation
MULTI uses two-stage textual inversion to disentangle camera lens, sensor, view, and domain factors for novel image generation, supporting dataset extension and ControlNet modifications on the new DF-RICO benchmark.
-
Revisiting Shadow Detection from a Vision-Language Perspective
SVL uses vision-language alignment via scene-level shadow ratio regression and global-to-local coupling on a frozen DINOv3 encoder to disambiguate shadows from dark surfaces in dense prediction.
-
MPerS: Dynamic MLLM MixExperts Perception-Guided Remote Sensing Scene Segmentation
MPerS dynamically mixes semantic guidance from MLLM-generated RS captions with DINOv3 features via MixExperts and Linguistic Query Guided Attention to achieve superior semantic segmentation on three public remote sensing datasets.
-
The Alpha Blending Hypothesis: Compositing Shortcut in Deepfake Detection
Deepfake detectors act as alpha blending searchers; training solely on self-blended real images yields top cross-dataset generalization on 15 datasets without using synthetic deepfakes.
-
Noise-Started One-Step Real-World Super-Resolution via LR-Conditioned SplitMeanFlow and GAN Refinement
SMFSR achieves state-of-the-art perceptual quality among one-step diffusion-based real-world super-resolution methods by preserving noise-started generation via LR-conditioned SplitMeanFlow and GAN refinement.
-
Unison: Harmonizing Motion, Speech, and Sound for Human-Centric Audio-Video Generation
Unison presents a unified audio-video generation model that decouples speech and sound effects while using bidirectional forcing to synchronize with motion, claiming SOTA perceptual quality and alignment.
-
Improving Generative Adversarial Networks with Self-Distillation
SD-GAN uses the EMA generator as a teacher to distill perceptual knowledge to the training generator, improving FID scores, stabilizing training, and providing guidance uncorrelated with standard adversarial loss.
-
ICDAR 2026 Competition on Writer Identification and Pen Classification from Hand-Drawn Circles
CircleID introduces a controlled dataset of 46,155 circles from 66 writers and 8 pens, with competition results showing top accuracies of 64.8% for open-set writer identification and 92.7% for pen classification.
-
Dynamic Mode Decomposition along Depth in Vision Transformers
Dynamic Mode Decomposition shows that short contiguous spans of Vision Transformer blocks can be approximated by a low-rank linear operator K with high predictive fidelity for p<=4 steps, but this approximation fails to outperform an identity baseline when propagated to the final layer.
-
D-OPSD: On-Policy Self-Distillation for Continuously Tuning Step-Distilled Diffusion Models
D-OPSD formulates supervised fine-tuning of step-distilled diffusion models as on-policy self-distillation by having the model act as both teacher (with multimodal context) and student (with text-only context) on its own roll-outs.
-
LaST-R1: Reinforcing Robotic Manipulation via Adaptive Physical Latent Reasoning
LaST-R1 introduces a RL post-training method called LAPO that optimizes latent Chain-of-Thought reasoning in vision-language-action models, yielding 99.9% success on LIBERO and up to 22.5% real-world gains.
-
Generalized Category Discovery under Domain Shifts: From Vision to Vision-Language Models
Three frameworks adapt foundation models for generalized category discovery under domain shifts via disentanglement and prompt tuning, showing gains on synthetic and real multi-domain data.
-
ICPR 2026 Competition on Low-Resolution License Plate Recognition
The ICPR 2026 LRLPR competition on real low-quality license plate images drew 99 valid submissions, with the winning team reaching 82.13% recognition rate and four teams exceeding 80%.
-
LSRM: High-Fidelity Object-Centric Reconstruction via Scaled Context Windows
Scaling transformer context with sparse attention and 3D-aware block routing improves feed-forward 3D reconstruction and inverse rendering, closing much of the quality gap with dense-view optimization.
-
Veo-Act: How Far Can Frontier Video Models Advance Generalizable Robot Manipulation?
Veo-3 video predictions enable approximate task-level robot trajectories in zero-shot settings but require hierarchical integration with low-level VLA policies for reliable manipulation performance.
-
CANDLE: Illumination-Invariant Semantic Priors for Color Ambient Lighting Normalization
CANDLE uses DINOv3 semantic features as illumination-invariant priors plus custom decoder refinements to normalize images under colored ambient lighting, reporting a 1.22 dB PSNR gain and strong challenge rankings.
-
DVGT-2: Vision-Geometry-Action Model for Autonomous Driving at Scale
DVGT-2 is a streaming vision-geometry-action model that jointly reconstructs dense 3D geometry and plans trajectories online, achieving better reconstruction than prior batch methods while transferring directly to planning benchmarks without fine-tuning.
-
EponaV2: Driving World Model with Comprehensive Future Reasoning
EponaV2 advances perception-free driving world models by forecasting comprehensive future 3D geometry and semantic representations, achieving SOTA planning performance on NAVSIM benchmarks.
-
DenseTRF: Texture-Aware Unsupervised Representation Adaptation for Surgical Scene Dense Prediction
DenseTRF adapts texture-aware representations via slot attention for unsupervised improvement of cross-domain generalization in surgical dense prediction tasks.
-
DetRefiner: Model-Agnostic Detection Refinement with Feature Fusion Transformer
DetRefiner fuses global and local features with a Transformer to refine OVOD confidence scores, delivering up to +10.1 AP gains on novel categories across multiple datasets.
-
VL-SAM-v3: Memory-Guided Visual Priors for Open-World Object Detection
VL-SAM-v3 retrieves visual prototypes from memory to generate sparse spatial and dense contextual priors that refine detection prompts, yielding gains on rare categories in LVIS for both open-vocabulary and open-ended settings.
-
VLA Foundry: A Unified Framework for Training Vision-Language-Action Models
VLA Foundry provides a single training stack for VLA models and releases open models that match prior closed-source performance or outperform baselines on multi-task manipulation in simulation.
-
SCC-Loc: A Unified Semantic Cascade Consensus Framework for UAV Thermal Geo-Localization
SCC-Loc claims SOTA UAV thermal geo-localization (9.37 m mean error, 7.6× better at 5 m) via a shared-DINOv2 semantic cascade consensus pipeline and a new Thermal-UAV dataset.
-
Position: Life-Logging Video Streams Make the Privacy-Utility Trade-off Inevitable
Life-logging video streams create an inevitable privacy-utility trade-off that is a foundational challenge for always-on AI systems.
-
Why Do DiT Editors Drift? Plug-and-Play Low Frequency Alignment in VAE Latent Space
VAE-LFA suppresses semantic drift in multi-turn DiT image editing by low-pass filtering latent discrepancies and aligning low-frequency components to an EMA of previous rounds in VAE space.
-
The Second Challenge on Cross-Domain Few-Shot Object Detection at NTIRE 2026: Methods and Results
The NTIRE 2026 CD-FSOD Challenge report details innovative methods and performance results from 19 teams on cross-domain few-shot object detection in open- and closed-source tracks.
-
Using Deep Learning Models Pretrained by Self-Supervised Learning for Protein Localization
DINO-based ViT models pretrained on HPA FOV achieve macro F1 of 0.822 zero-shot and 0.860 after fine-tuning for protein localization on OpenCell, demonstrating effective transfer from SSL pretraining.