$\mathrm{D}^\mathrm{3}$-Predictor: Noise-Free Deterministic Diffusion for Dense Prediction

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Xia, Changliang, Jia, Chengyou, Luo, Minnan, Dang, Zhuohang, Shen, Xin, Ping, Bowen
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866912836340416512
author Xia, Changliang
Jia, Chengyou
Luo, Minnan
Dang, Zhuohang
Shen, Xin
Ping, Bowen
author_facet Xia, Changliang
Jia, Chengyou
Luo, Minnan
Dang, Zhuohang
Shen, Xin
Ping, Bowen
contents Although diffusion models with strong visual priors have emerged as powerful dense prediction backbones, they overlook a core limitation: the stochastic noise at the core of diffusion sampling is inherently misaligned with dense prediction that requires a deterministic mapping from image to geometry. In this paper, we show that this stochastic noise corrupts fine-grained spatial cues and pushes the model toward timestep-specific noise objectives, consequently destroying meaningful geometric structure mappings. To address this, we introduce $\mathrm{D}^\mathrm{3}$-Predictor, a noise-free deterministic diffusion-based dense prediction model built by reformulating a pretrained diffusion model without stochasticity noise. Instead of relying on noisy inputs to leverage diffusion priors, $\mathrm{D}^\mathrm{3}$-Predictor views the pretrained diffusion network as an ensemble of timestep-dependent visual experts and self-supervisedly aggregates their heterogeneous priors into a single, clean, and complete geometric prior. Meanwhile, we utilize task-specific supervision to seamlessly adapt this noise-free prior to dense prediction tasks. Extensive experiments on various dense prediction tasks demonstrate that $\mathrm{D}^\mathrm{3}$-Predictor achieves competitive or state-of-the-art performance in diverse scenarios. In addition, it requires less than half the training data previously used and efficiently performs inference in a single step. Our code, data, and checkpoints are publicly available at https://x-gengroup.github.io/HomePage_D3-Predictor/.
format Preprint
id arxiv_https___arxiv_org_abs_2512_07062
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle $\mathrm{D}^\mathrm{3}$-Predictor: Noise-Free Deterministic Diffusion for Dense Prediction
Xia, Changliang
Jia, Chengyou
Luo, Minnan
Dang, Zhuohang
Shen, Xin
Ping, Bowen
Computer Vision and Pattern Recognition
Artificial Intelligence
Although diffusion models with strong visual priors have emerged as powerful dense prediction backbones, they overlook a core limitation: the stochastic noise at the core of diffusion sampling is inherently misaligned with dense prediction that requires a deterministic mapping from image to geometry. In this paper, we show that this stochastic noise corrupts fine-grained spatial cues and pushes the model toward timestep-specific noise objectives, consequently destroying meaningful geometric structure mappings. To address this, we introduce $\mathrm{D}^\mathrm{3}$-Predictor, a noise-free deterministic diffusion-based dense prediction model built by reformulating a pretrained diffusion model without stochasticity noise. Instead of relying on noisy inputs to leverage diffusion priors, $\mathrm{D}^\mathrm{3}$-Predictor views the pretrained diffusion network as an ensemble of timestep-dependent visual experts and self-supervisedly aggregates their heterogeneous priors into a single, clean, and complete geometric prior. Meanwhile, we utilize task-specific supervision to seamlessly adapt this noise-free prior to dense prediction tasks. Extensive experiments on various dense prediction tasks demonstrate that $\mathrm{D}^\mathrm{3}$-Predictor achieves competitive or state-of-the-art performance in diverse scenarios. In addition, it requires less than half the training data previously used and efficiently performs inference in a single step. Our code, data, and checkpoints are publicly available at https://x-gengroup.github.io/HomePage_D3-Predictor/.
title $\mathrm{D}^\mathrm{3}$-Predictor: Noise-Free Deterministic Diffusion for Dense Prediction
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2512.07062