TurboClear

One-step object-effect removal

TurboClear

One-Step Object-Effect Removal via Region-Calibrated Distribution Matching and Fusion

Jiawei Guo1, Junxian Li1, Yixin Tang1, Bingya Zhang2, Jiaxin Lu2, Yulun Zhang1,†, Shangchen Zhou3,†

1 Shanghai Jiao Tong University 2 Honor Device Co., Ltd 3 Imperial College London

† Corresponding authors

Remove the target object, its shadow, reflection, and other visual effects in a single denoising step while keeping unaffected regions intact.

Paper PDF Model Weights Project
Calm lake after TurboClear removes a sailboat and its reflection
Sailboat selected with a green mask on a calm lake
Input + mask TurboClear
Waterfall after TurboClear removes a person and their shadow
Person selected with a green mask in front of a waterfall
Input + mask TurboClear
Basketball court after TurboClear removes a foreground player and their shadow
Foreground basketball player selected with a green mask
Input + mask TurboClear

Drag each divider to inspect the input and result.

1denoising step
1.589 Tdenoising FLOPs on OBER-Test
40.04×less computation than ObjectClear
665×less computation than OmniPaint

Why TurboClear

Edit what matters. Preserve what does not.

Object removal must reconstruct areas influenced by the target while leaving unrelated content unchanged. Generic one-step objectives do not explicitly model this spatial asymmetry, which can leave residual effects or introduce unnecessary background changes.

TurboClear couples Region-Calibrated Distribution Matching with Learnable Spatial Fusion. The resulting SDXL-based student preserves the teacher's asymmetric edit-and-preserve behavior while reducing iterative diffusion inference to one step.

Qualitative results

Objects disappear. Structure stays.

Five OBER-Wild examples from the updated qualitative comparison in the paper.

Reflective water after TurboClear removes an island and its reflection
Island selected with a green mask against a sunset sky
Input + mask TurboClear
Sunset over water after TurboClear removes a boat and its reflection
Boat selected with a green mask on water at sunset
Input + mask TurboClear
Full qualitative comparison of TurboClear and prior object removal methods
Comparison on five representative OBER-Wild samples.

Method

Region-aware distillation, spatially learned fusion

TurboClear training and inference pipeline with RDM and LSF
RDM calibrates distribution matching across affected and preserved regions during training. At inference, LSF predicts latent- and pixel-space gates to combine the generated output with the input stream.

Efficiency

Quality without the iterative cost

TurboClear reaches the strongest PSNR in the comparison while using substantially fewer denoising FLOPs than multi-step diffusion removal systems. Bubble area denotes model parameter count.

PSNR versus denoising FLOPs and parameter count for object removal methods

Citation

BibTeX

Please cite TurboClear using the following arXiv entry.

@article{guo2026turboclear,
  title={TurboClear: One-Step Object-Effect Removal via Region-Calibrated Distribution Matching and Fusion},
  author={Guo, Jiawei and Li, Junxian and Tang, Yixin and Zhang, Bingya and Lu, Jiaxin and Zhang, Yulun and Zhou, Shangchen},
  journal={arXiv preprint arXiv:2608.01288},
  year={2026}
}

Resources

Release status

Paper
arXiv:2608.01288
Model weights
Hugging Face / JGuo666/TurboClear
Code
GitHub / GuoCalix/TurboClear
Usage guide
Inference and training README
License
Apache License 2.0
Acknowledgements
Inspired by ObjectClear and DMD2