BOCCHI: A More Realistic and Challenging Benchmark for Local Motion Blur Detection with MSDCT-UNet

0.850

In-domain mIoU on BOCCHI

First among 13 methods, +1.2 pp over the second-best baseline.

633

Training images

Enough to beat every other training source on cross-dataset transfer.

3

of 3 cross-dataset metrics

BOCCHI-trained models lead on mIoU, Dice and Recall alike.

Abstract

Fewer images, better transfer

Local motion blur detection requires pixel-level localization of blurred regions. Existing benchmarks let models rely on gradient shortcuts that fail to transfer. We introduce BOCCHI (Blurred Objects Captured across Cameras with Human-annotated Imagery), a real-captured benchmark of 633 pixel-annotated images whose textured blurred foregrounds against varied backgrounds defeat these shortcuts, and propose MSDCT-UNet (Multi-Scale Discrete Cosine Transform UNet), a frequency-aware encoder-decoder injecting multi-scale DCT priors through DCT Attention and FiLM. MSDCT-UNet ranks first in in-domain mIoU (0.850, +1.2 pp over the second-best baseline) and boundary localization on BOCCHI. BOCCHI-trained models outperform every other training source on cross-dataset transfer with only 633 training images: +2.2 pp mIoU over OMoBlur, +13.8 pp over ReLoBlur, and +34.9 pp over CUHK_motion.

Benchmark

The BOCCHI dataset

633 real-captured images across 5 cameras and diverse indoor/outdoor scenes, each annotated at pixel level.

BOCCHI

Inference Set

633

Images

5

Cameras

0.22

Mean blur ratio

68.7

Mean sharp gradient

Two distinguishing properties

Blur-ratio distribution and gradient statistics across datasets

(a) Blur-ratio distribution across 5 datasets: CUHK_motion spans uniformly; ReLoBlur and BOCCHI concentrate at low ratios; OMoBlur occupies the moderate range; the Inference Dataset spans the intermediate regime. (b) Per-dataset distribution of mean gradient in blur vs. sharp regions. BOCCHI shows the largest PR25/μ_blur ratio (0.68): the sharp-region bottom-25% gradient overlaps the blur mean, defeating gradient-based shortcuts.

Data efficiency

Which training source transfers best?

Same 13 models, same held-out Inference Dataset, no fine-tuning. Only the training source changes.

BOCCHI

633 images

0.000

OMoBlur

994 images

0.000

ReLoBlur

1,200 images

0.000

CUHK_motion

200 images

0.000

Cross-dataset mIoU, averaged over 13 models (0 – 0.60)

More training images do not buy better transfer: ReLoBlur has almost twice BOCCHI's images and lands 13.8 pp lower. The OMoBlur average aggregates 12 of 13 models (NAFNet × OMoBlur omitted).

Qualitative

Predictions, method by method

Switch methods to compare predictions on the BOCCHI validation set. Each panel shows 4 input–prediction pairs.

Currently showing GT reference

Selected success cases where MSDCT-UNet outperforms baselines. Predictions are binary masks (white = sharp, black = predicted blur).

Quantitative

Results

In-domain performance

Per-cell metrics: mIoU / BdF1 on each dataset's own validation set. Cells shaded 1st 2nd 3rd per column. ^*NAFNet × OMoBlur omitted (aspect-ratio width collapse).

Method	BOCCHI		CUHK_motion		ReLoBlur		OMoBlur
Method	mIoU	BdF1	mIoU	BdF1	mIoU	BdF1	mIoU	BdF1
MSDU-Net	0.796	0.561	0.640	0.330	0.923	0.820	0.772	0.667
BiSeNetV2	0.760	0.470	0.525	0.166	0.916	0.800	0.749	0.613
STDC1	0.778	0.457	0.550	0.260	0.918	0.795	0.748	0.631
STDC2	0.799	0.495	0.558	0.197	0.918	0.800	0.755	0.627
NAFNet	0.760	0.452	0.523	0.213	0.914	0.821	N/A^*	N/A^*
DDRNet-23	0.701	0.351	0.520	0.172	0.910	0.766	0.703	0.583
Cellpose3	0.838	0.643	0.665	0.403	0.927	0.850	0.791	0.675
KDSNet-R50	0.783	0.469	0.549	0.219	0.918	0.803	0.748	0.627
KDSNet-R101	0.773	0.446	0.544	0.243	0.921	0.812	0.744	0.630
MSDSeg	0.714	0.360	0.517	0.202	0.909	0.766	0.721	0.607
BEVANet	0.778	0.518	0.548	0.217	0.919	0.798	0.721	0.622
ESMDL-UNet	0.677	0.270	0.432	0.139	0.788	0.364	0.627	0.552
MSDCT-UNet (Ours)	0.850	0.683	0.640	0.369	0.930	0.851	0.775	0.664

Cross-dataset generalization

mIoU / Dice / Recall on the Inference Dataset, no fine-tuning. The average row highlights BOCCHI as the strongest training source. ^*NAFNet × OMoBlur omitted (OMoBlur average aggregates 12 of 13).

Method	Trained on BOCCHI	Trained on OMoBlur	Trained on ReLoBlur	Trained on CUHK_motion
Method	mIoU / Dice / Rec	mIoU / Dice / Rec	mIoU / Dice / Rec	mIoU / Dice / Rec
MSDU-Net	0.609 / 0.830 / 0.876	0.616 / 0.805 / 0.810	0.556 / 0.758 / 0.736	0.142 / 0.307 / 0.268
BiSeNetV2	0.539 / 0.791 / 0.843	0.526 / 0.704 / 0.666	0.470 / 0.695 / 0.651	0.248 / 0.445 / 0.455
STDC1	0.541 / 0.789 / 0.840	0.496 / 0.657 / 0.597	0.283 / 0.518 / 0.538	0.197 / 0.310 / 0.286
STDC2	0.577 / 0.823 / 0.882	0.530 / 0.705 / 0.655	0.269 / 0.443 / 0.437	0.182 / 0.288 / 0.277
NAFNet	0.568 / 0.818 / 0.882	N/A^*	0.478 / 0.730 / 0.718	0.285 / 0.529 / 0.533
DDRNet-23	0.517 / 0.800 / 0.877	0.489 / 0.673 / 0.626	0.462 / 0.704 / 0.665	0.292 / 0.548 / 0.490
Cellpose3	0.662 / 0.858 / 0.910	0.628 / 0.825 / 0.848	0.555 / 0.758 / 0.744	0.160 / 0.324 / 0.306
KDSNet-R50	0.557 / 0.809 / 0.871	0.528 / 0.717 / 0.684	0.275 / 0.465 / 0.465	0.225 / 0.424 / 0.369
KDSNet-R101	0.557 / 0.794 / 0.835	0.525 / 0.718 / 0.696	0.274 / 0.433 / 0.423	0.230 / 0.432 / 0.372
MSDSeg	0.518 / 0.816 / 0.901	0.547 / 0.788 / 0.828	0.463 / 0.726 / 0.711	0.141 / 0.128 / 0.121
BEVANet	0.552 / 0.804 / 0.861	0.535 / 0.746 / 0.731	0.484 / 0.718 / 0.692	0.271 / 0.493 / 0.434
ESMDL-UNet	0.518 / 0.806 / 0.873	0.458 / 0.698 / 0.684	0.391 / 0.631 / 0.593	0.231 / 0.397 / 0.352
MSDCT-UNet (Ours)	0.610 / 0.848 / 0.925	0.610 / 0.783 / 0.781	0.563 / 0.815 / 0.897	0.183 / 0.290 / 0.226
Average	*0.563 / 0.814 / 0.875*	0.541 / 0.735 / 0.717^*	0.425 / 0.646 / 0.636	0.214 / 0.378 / 0.345

Per-image comparison

Each dot is one validation image (n = 63). x = mIoU of the strongest baseline on that image; y = mIoU of MSDCT-UNet. Above the diagonal we win, below the baseline wins. Highlighted dots are the 8 cases in the qualitative panel; click one to jump there.

Ours wins Baseline wins Shown above (click to jump)

Method

MSDCT-UNet

Left: DCT feature extraction (grayscale + Sobel → multi-scale local DCTs at n ∈ {3, 7, 15, 31} → 57-channel F_DCT). Middle: 4-stage encoder-decoder with NeXtBlock + FiLM fusion, an Attentive Frequency ASPP bottleneck, and stage-specific DCT Attention. Right: Deep supervision over the main and three auxiliary heads.

Key modules

DCT Attention

Multi-head, position-dependent frequency weighting over 57 DCT channels with SE-gating and temperature-scaled softmax.

FiLM Fusion

Frequency-conditioned affine modulation prevents strong spatial edges from overriding weaker frequency cues.

AFASPP

Attentive Frequency ASPP fuses multi-scale spatial context with global frequency evidence at the bottleneck.

Full method details, including DCT extraction pseudocode and ablations, are available in the paper.