DiTGaze：几何条件扩散增强实现少样本视线估计

View PDF HTML (experimental)

Abstract:Accurate person-specific gaze estimation benefits from labeled user data, but collecting images that densely cover gaze directions is burdensome. We present DiTGaze, a calibration-efficient augmentation system that follows a two-stream 3D Gaussian Splatting backbone and reformulates its convolutional rendering stage as a geometry-conditioned latent Diffusion Transformer (DiT). A learnable projection and frozen VAE bridge rasterized Gaussian features to latent tokens, while timestep and identity/expression conditioning are integrated through Adaptive Layer Normalization. Intermediate yaw-pitch sampling densifies gaze supervision, and an orthogonality constraint reduces interference between gaze, pose, and expression. On ETH-XGaze, DiTGaze reduces generator-level gaze error by 4.1 percent relative to GazeGaussian. In our runs of the expanded downstream protocol, it ranks first at every calibration budget from one to nine real images; the repeated primary comparison at N = 9 reduces subject-level estimator error by 0.73 degrees and improves 14 of 15 identities (p < 0.001). These results connect the proposed generator changes to statistically supported few-shot personalization gains.

Comments:	Accepted at WACV 2027. Updated and extended version of the original 2025 preprint, including revised methodology and expanded few-shot gaze-estimation experiments
Subjects:	Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Cite as:	arXiv:2511.11231 [cs.CV]
	(or arXiv:2511.11231v2 [cs.CV] for this version)
	https://doi.org/10.48550/arXiv.2511.11231 arXiv-issued DOI via DataCite

Submission history

From: Abiram Panchalingam Mr [view email]
[v1] Fri, 14 Nov 2025 12:32:22 UTC (688 KB)
[v2] Thu, 29 Oct 2026 19:17:00 UTC (1,930 KB)