Track4World: Feedforward World-centric Dense 3D Tracking of All Pixels

Jiahao Lu; Jiayi Xu; Wenbo Hu; Ruijie Zhu; Chengfeng Zhao; Sai-Kit Yeung; Ying Shan; Yuan Liu

cs.CVMar 3, 2026

Track4World: Feedforward World-centric Dense 3D Tracking of All Pixels

Authors: Jiahao Lu, Jiayi Xu, Wenbo Hu, Ruijie Zhu, Chengfeng Zhao, Sai-Kit Yeung, Ying Shan, Yuan Liu

Organizations: The Hong Kong University of Science and Technology · ARC Lab, Tencent PCG

Abstract

Estimating the 3D trajectory of every pixel from a monocular video is crucial and promising for a comprehensive understanding of the 3D dynamics of videos. Recent monocular 3D tracking works demonstrate impressive performance, but are limited to either tracking sparse points on the first frame or a slow optimization-based framework for dense tracking. In this paper, we propose a feedforward model, called Track4World, enabling an efficient holistic 3D tracking of every pixel in the world-centric coordinate system. Built on the global 3D scene representation encoded by a VGGT-style ViT, Track4World applies a novel 3D correlation scheme to simultaneously estimate the pixel-wise 2D and 3D dense flow between arbitrary frame pairs. The estimated scene flow, along with the reconstructed 3D geometry, enables subsequent efficient 3D tracking of every pixel of this video. Extensive experiments on multiple benchmarks demonstrate that our approach consistently outperforms existing methods in 2D/3D flow estimation and 3D tracking, highlighting its robustness and scalability for real-world 4D reconstruction tasks.

Figures & tables

Figure 2: Overview . Given (a) the input video frames, Track4World first extracts (b) global scene representations (geometric embeddings, point clouds, and camera poses). (c) A sparse-to-dense scene flow decoder then predicts 2D-3D joint flows between arbitrary timesteps, which applies a novel 2D-to-3D correlation scheme to improve efficiency and allows 2D-3D joint supervision. (d) The pairwise flows are ultimately fused to establish holistic world-centric 3D tracking.

Figure 3: Comparison of correlation mechanisms. Prior methods rely on explicit $k$ -nearest neighbor searches and cross-attention in 3D space, leading to high computational costs. In contrast, our proposed method anchors 3D updates directly to intermediate image-plane correlations. This design significantly improves computational efficiency and allows the 3D tracking module to be effectively boosted by abundant 2D training data.

Camera coordinate 3D tracking
Method	PointOdyssey [ 102 ]		ADT [ 61 ]		PStudio [ 32 ]		DriveTrack [ 3 ]		Avg.
Method	L-16	L-50	L-16	L-50	L-16	L-50	L-16	L-50	L-16	L-50
SpatialTracker ∗ [ 87 ]	0.3116	0.2977	0.4962	0.4692	0.5390	0.4991	0.2529	0.2502	0.3999	0.3791
DELTA ∗ [ 59 ]	0.3529	0.3412	0.5116	0.4952	0.5922	0.5533	0.2704	0.2701	0.4317	0.4150
STV2 † [ 86 ]	0.1864	0.1785	0.2400	0.2330	0.3784	0.3690	0.1711	0.1725	0.2400	0.2383
MASt3R [ 43 ]	0.3546	0.3253	0.3368	0.3029	0.3293	0.2956	0.2767	0.2559	0.3244	0.2949

Table 2: 3D tracking estimation. We report the APD metric to evaluate 3D point tracking on the PointOdyssey [ 102 ] , ADT [ 61 ] , PStudio [ 32 ] , and DriveTrack [ 3 ] datasets. L-16 and L-50 indicate tracking within the temporal length of 16 and 50 frames, respectively. * indicates that ground-truth camera intrinsics are used.

\dagger

indicates that bundle adjustment is not used. ‡ denotes methods using camera poses estimated by VGGT [ 76 ] .

Figure 4: Qualitative results on diverse in-the-wild videos.

Appendix figures & tables8 assets

Supplementary material from the paper’s appendix.

Appendix

Fig. S1 : Effectiveness of $\ell_{\text{smooth}}^{3d}$ .

Setting	Abs Rel $\downarrow$	$\delta<1.25\uparrow$
w/ scale-inv	0.1060	0.8788
w/o $\ell_{reg}$	0.3530	0.3547
w/o $\ell_{local}$	0.1053	0.8790
Full (Ours)	0.0973	0.8921

Appendix

Table S1 : Ablation study on geometry encoder. The results are averaged over all datasets. Best results are highlighted in darker blue , and second best in lighter blue .

Fig. S2 : Scene flow Visualization. The deformed point maps (colored ) show that our method produces more temporally consistent geometry and motion.

Fig. S3 : Scene Flow Visualization. The deformed point maps (shown in solid colors) indicate that our method produces higher-quality geometry and motion.

Fig. S4 : Visualization of first-frame 2D dense tracking.

Fig. S5 : Visualization of first-frame 3D dense tracking.

Fig. S6 : Visualization of dense per-pixel trajectories across all frames.

Fig. S7 : Visualization of the dense trajectories across all frames in the global world coordinate system.

Scene Flow Estimation

See all 12 papers

TrackCraft3R: Repurposing Video Diffusion Transformers for Dense 3D Tracking
May 12, 2026Jisu Nam, Jahyeok Koo, Soowon Son +4Point Tracking 3D Point Tracking
KAIST AI · Google DeepMind
MoSE3: Learning World-Space SE(3) at Every Pixel
Oct 2, 2026Jiahuan Cheng, Zhiyi Li, Tian Xia +33D Point Tracking
Harvard University · Johns Hopkins University · MIT +1
TrackEverything: Long Horizon Dense Tracking via De-Duplicating 3D Scene Representations
Sep 24, 2026Ayush Jain, Sreeharsha Paruchuri, Ishita Gupta +5Point Tracking 3D Point Tracking
Carnegie Mellon University · Meta

May 12, 2026J/K move · Enter open · S save

TrackCraft3R: Repurposing Video Diffusion Transformers for Dense 3D Tracking

Jisu Nam, Jahyeok Koo, Soowon Son, Jaewoo Jung+3

KAIST AI · Google DeepMind

Point Tracking 3D Point Tracking Video Diffusion Models Visual Object Tracking

Open paper AskPDF arXiv

Dense 3D tracking from monocular video is fundamental to dynamic scene understanding. While recent 3D foundation models provide reliable per-frame geometry, recovering object motion in this geometry remains challenging and benefits from strong motion priors learned from real-world videos. Existing 3D trackers either follow iterative paradigms trained from scratch on synthetic data or fine-tune 3D reconstruction models learned from static multi-view images, both lacking real-world motion priors. Pre-trained video diffusion transformers (video DiTs) offer rich spatio-temporal priors from internet-scale videos, making them a promising foundation for 3D tracking. However, their frame-anchored formulation, which generates each frame's content, is fundamentally mismatched with reference-anchored dense 3D tracking, which must follow the same physical points from a reference frame across time. We present TrackCraft3R, the first method to repurpose a video DiT as a feed-forward dense 3D tracker. Given a monocular video and its frame-anchored reconstruction pointmap, TrackCraft3R predicts a reference-anchored tracking pointmap that follows every pixel of the first frame across time in a single forward pass, along with its visibility. We achieve this through two designs: (i) a dual-latent representation that uses per-frame geometry latents and reference-anchored track latents as dense queries, and (ii) temporal RoPE alignment, which specifies the target timestamp of each track latent. Together, these designs convert the per-frame generative paradigm of video DiTs into a reference-anchored tracking formulation with LoRA fine-tuning. TrackCraft3R achieves state-of-the-art performance on standard sparse and dense 3D tracking benchmarks, while running 1.3x faster and using 4.6x less peak memory than the strongest prior method. We further demonstrate robustness to large motions and long videos.

Abstract

Scene Flow Estimation

Explore similar work

TrackCraft3R: Repurposing Video Diffusion Transformers for Dense 3D Tracking

MoSE3: Learning World-Space SE(3) at Every Pixel

TrackEverything: Long Horizon Dense Tracking via De-Duplicating 3D Scene Representations