Benchmarking Dense and Indiscernible Object Counting with Blueberries

Weihao Bo, Yanpeng Sun, Jingwen Qin, Fei Shen, Xiaofan Li, Zechao Li
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:8676-8691, 2026.

Abstract

Real-world agricultural counting often operates in the extreme regime of Dense and Indiscernible Object Counting (DIOC), where targets are tiny, clustered, and highly camouflaged. To facilitate research in this domain, we introduce DIOCblueberry, a large-scale benchmark that pushes the boundaries of visual perception. Unlike general datasets with salient objects, DIOCblueberry features extreme occlusion and camouflage. Compared to the popular FSC147 benchmark, it contains 1.9$\times$ more instances per image (avg. 108) with an average box pixel ratio that is 7.9$\times$ smaller, serving as a rigorous testbed for model robustness. Standard counting methods struggle in these scenarios due to severe visual ambiguity and scale mismatch. To address this, we propose MaskCount, a coarse-to-fine framework that incorporates semantic guidance. MaskCount leverages Vision-Language Models (CLIP) to generate pseudo segmentation masks for background suppression and employs a contrastive loss to maximize feature discriminability between fruits and foliage. Additionally, we design an edge-aware cropping mechanism to resolve boundary truncation in dense clusters. Extensive experiments demonstrate that MaskCount achieves a new state-of-the-art, reducing MAE and RMSE by 49.16% and 70.50% respectively on DIOCblueberry, with strong generalization to other agricultural scenes.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-bo26a, title = {Benchmarking Dense and Indiscernible Object Counting with Blueberries}, author = {Bo, Weihao and Sun, Yanpeng and Qin, Jingwen and Shen, Fei and Li, Xiaofan and Li, Zechao}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {8676--8691}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/bo26a/bo26a.pdf}, url = {https://proceedings.mlr.press/v306/bo26a.html}, abstract = {Real-world agricultural counting often operates in the extreme regime of Dense and Indiscernible Object Counting (DIOC), where targets are tiny, clustered, and highly camouflaged. To facilitate research in this domain, we introduce DIOCblueberry, a large-scale benchmark that pushes the boundaries of visual perception. Unlike general datasets with salient objects, DIOCblueberry features extreme occlusion and camouflage. Compared to the popular FSC147 benchmark, it contains 1.9$\times$ more instances per image (avg. 108) with an average box pixel ratio that is 7.9$\times$ smaller, serving as a rigorous testbed for model robustness. Standard counting methods struggle in these scenarios due to severe visual ambiguity and scale mismatch. To address this, we propose MaskCount, a coarse-to-fine framework that incorporates semantic guidance. MaskCount leverages Vision-Language Models (CLIP) to generate pseudo segmentation masks for background suppression and employs a contrastive loss to maximize feature discriminability between fruits and foliage. Additionally, we design an edge-aware cropping mechanism to resolve boundary truncation in dense clusters. Extensive experiments demonstrate that MaskCount achieves a new state-of-the-art, reducing MAE and RMSE by 49.16% and 70.50% respectively on DIOCblueberry, with strong generalization to other agricultural scenes.} }
Endnote
%0 Conference Paper %T Benchmarking Dense and Indiscernible Object Counting with Blueberries %A Weihao Bo %A Yanpeng Sun %A Jingwen Qin %A Fei Shen %A Xiaofan Li %A Zechao Li %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-bo26a %I PMLR %P 8676--8691 %U https://proceedings.mlr.press/v306/bo26a.html %V 306 %X Real-world agricultural counting often operates in the extreme regime of Dense and Indiscernible Object Counting (DIOC), where targets are tiny, clustered, and highly camouflaged. To facilitate research in this domain, we introduce DIOCblueberry, a large-scale benchmark that pushes the boundaries of visual perception. Unlike general datasets with salient objects, DIOCblueberry features extreme occlusion and camouflage. Compared to the popular FSC147 benchmark, it contains 1.9$\times$ more instances per image (avg. 108) with an average box pixel ratio that is 7.9$\times$ smaller, serving as a rigorous testbed for model robustness. Standard counting methods struggle in these scenarios due to severe visual ambiguity and scale mismatch. To address this, we propose MaskCount, a coarse-to-fine framework that incorporates semantic guidance. MaskCount leverages Vision-Language Models (CLIP) to generate pseudo segmentation masks for background suppression and employs a contrastive loss to maximize feature discriminability between fruits and foliage. Additionally, we design an edge-aware cropping mechanism to resolve boundary truncation in dense clusters. Extensive experiments demonstrate that MaskCount achieves a new state-of-the-art, reducing MAE and RMSE by 49.16% and 70.50% respectively on DIOCblueberry, with strong generalization to other agricultural scenes.
APA
Bo, W., Sun, Y., Qin, J., Shen, F., Li, X. & Li, Z.. (2026). Benchmarking Dense and Indiscernible Object Counting with Blueberries. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:8676-8691 Available from https://proceedings.mlr.press/v306/bo26a.html.

Related Material