# DeNet: Scalable Real-time Object Detection with Directed Sparse Sampling

arXiv: https://arxiv.org/abs/1703.10295 · HTML: https://mlsanity.com/paper/1703.10295/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 30 Mar 2017.

## Built on by (most cited first, 12 of 12)

- [Recent advances and clinical applications of deep learning in medical image analysis](https://mlsanity.com/paper/2105.13381/) (2021)
- [Back-tracing Representative Points for Voting-based 3D Object Detection in Point Clouds](https://mlsanity.com/paper/2104.06114/) (2021)
- [ObjectBox: From Centers to Boxes for Anchor-Free Object Detection](https://mlsanity.com/paper/2207.06985/) (2022)
- [Continuous Human Action Recognition for Human-Machine Interaction: A Review](https://mlsanity.com/paper/2202.13096/) (2022)
- [CenterNet++ for Object Detection](https://mlsanity.com/paper/2204.08394/) (2022)
- [Location-Sensitive Visual Recognition with Cross-IOU Loss](https://mlsanity.com/paper/2104.04899/) (2021)
- [A Survey on Deep Domain Adaptation and Tiny Object Detection Challenges, Techniques and Datasets](https://mlsanity.com/paper/2107.07927/) (2021)
- [Decoupled IoU Regression for Object Detection](https://mlsanity.com/paper/2202.00866/) (2022)
- [R(Det)^2: Randomized Decision Routing for Object Detection](https://mlsanity.com/paper/2204.00794/) (2022)
- [Early Recall, Late Precision: Multi-Robot Semantic Object Mapping under Operational Constraints in Perceptually-Degraded Environments](https://mlsanity.com/paper/2206.10062/) (2022)
- [Pair DETR: Contrastive Learning Speeds Up DETR Training](https://mlsanity.com/paper/2210.16476/) (2022)
- [Autonomous 3D Visual-Thermal Inspection for Utility-Scale Solar Farms](https://mlsanity.com/paper/2610.17762/) (2026)

## Abstract (arXiv)

We define the object detection from imagery problem as estimating a very large but extremely sparse bounding box dependent probability distribution. Subsequently we identify a sparse distribution estimation scheme, Directed Sparse Sampling, and employ it in a single end-to-end CNN based detection model. This methodology extends and formalizes previous state-of-the-art detection models with an additional emphasis on high evaluation rates and reduced manual engineering. We introduce two novelties, a corner based region-of-interest estimator and a deconvolution based CNN model. The resulting model is scene adaptive, does not require manually defined reference bounding boxes and produces highly competitive results on MSCOCO, Pascal VOC 2007 and Pascal VOC 2012 with real-time evaluation rates. Further analysis suggests our model performs particularly well when finegrained object localization is desirable. We argue that this advantage stems from the significantly larger set of available regions-of-interest relative to other methods. Source-code is available from: https://github.com/lachlants/denet