# Knowledge Combination to Learn Rotated Detection Without Rotated Annotation

arXiv: https://arxiv.org/abs/2304.02199 · HTML: https://mlsanity.com/paper/2304.02199/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 38 references that are in the map.
- First posted on arXiv 5 Apr 2023.

## Builds on (most cited first, 23 of 23)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs](https://mlsanity.com/paper/1606.00915/) (2016)
- [Deep Learning and the Information Bottleneck Principle](https://mlsanity.com/paper/1503.02406/) (2015)
- [DOTA: A Large-scale Dataset for Object Detection in Aerial Images](https://mlsanity.com/paper/1711.10398/) (2017)
- [The Open Images Dataset V4: Unified image classification, object detection, and visual relationship detection at scale](https://mlsanity.com/paper/1811.00982/) (2018)
- [MOT16: A Benchmark for Multi-Object Tracking](https://mlsanity.com/paper/1603.00831/) (2016)
- [MOT20: A benchmark for multi object tracking in crowded scenes](https://mlsanity.com/paper/2003.09003/) (2020)
- [Learning Pixel-level Semantic Affinity with Image-level Supervision for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/1803.10464/) (2018)
- [Self-supervised Equivariant Attention Mechanism for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/2004.04581/) (2020)
- [BoxSup: Exploiting Bounding Boxes to Supervise Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1503.01640/) (2015)
- [Oriented R-CNN for Object Detection](https://mlsanity.com/paper/2108.05699/) (2021)
- [Seed, Expand and Constrain: Three Principles for Weakly-Supervised Image Segmentation](https://mlsanity.com/paper/1603.06098/) (2016)
- [Weakly- and Semi-Supervised Learning of a DCNN for Semantic Image Segmentation](https://mlsanity.com/paper/1502.02734/) (2015)
- [BoxInst: High-Performance Instance Segmentation with Box Annotations](https://mlsanity.com/paper/2012.02310/) (2020)
- [Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi- Supervised Semantic Segmentation](https://mlsanity.com/paper/1805.04574/) (2018)
- [Learning High-Precision Bounding Box for Rotated Object Detection via Kullback-Leibler Divergence](https://mlsanity.com/paper/2106.01883/) (2021)
- [MMRotate: A Rotated Object Detection Benchmark using PyTorch](https://mlsanity.com/paper/2204.13317/) (2022)
- [Weakly Supervised Object Localization and Detection: A Survey](https://mlsanity.com/paper/2104.07918/) (2021)
- [Instance-aware, Context-focused, and Memory-efficient Weakly Supervised Object Detection](https://mlsanity.com/paper/2004.04725/) (2020)

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

- [H2RBox-v2: Incorporating Symmetry for Boosting Horizontal Box Supervised Oriented Object Detection](https://mlsanity.com/paper/2304.04403/) (2023)
- [PointOBB: Learning Oriented Object Detection via Single Point Supervision](https://mlsanity.com/paper/2311.14757/) (2023)
- [Point2RBox: Combine Knowledge from Synthetic Visual Patterns for End-to-end Oriented Object Detection with Single Point Supervision](https://mlsanity.com/paper/2311.14758/) (2023)
- [PointOBB-v2: Towards Simpler, Faster, and Stronger Single Point Supervised Oriented Object Detection](https://mlsanity.com/paper/2410.08210/) (2024)
- [Point2RBox-v2: Rethinking Point-supervised Oriented Object Detection with Spatial Layout Among Instances](https://mlsanity.com/paper/2502.04268/) (2025)
- [P2RBox: Point Prompt Oriented Object Detection with SAM](https://mlsanity.com/paper/2311.13128/) (2023)
- [PointOBB-v3: Expanding Performance Boundaries of Single Point-Supervised Oriented Object Detection](https://mlsanity.com/paper/2501.13898/) (2025)
- [Wholly-WOOD: Wholly Leveraging Diversified-quality Labels for Weakly-supervised Oriented Object Detection](https://mlsanity.com/paper/2502.09471/) (2025)
- [Partial Weakly-Supervised Oriented Object Detection](https://mlsanity.com/paper/2507.02751/) (2025)
- [GOOD: Towards Domain Generalized Orientated Object Detection](https://mlsanity.com/paper/2402.12765/) (2024)
- [S2Teacher: Step-by-step Teacher for Sparsely Annotated Oriented Object Detection](https://mlsanity.com/paper/2504.11111/) (2025)
- [SPWOOD: Sparse Partial Weakly-Supervised Oriented Object Detection](https://mlsanity.com/paper/2602.03634/) (2026)

## Abstract (arXiv)

Rotated bounding boxes drastically reduce output ambiguity of elongated objects, making it superior to axis-aligned bounding boxes. Despite the effectiveness, rotated detectors are not widely employed. Annotating rotated bounding boxes is such a laborious process that they are not provided in many detection datasets where axis-aligned annotations are used instead. In this paper, we propose a framework that allows the model to predict precise rotated boxes only requiring cheaper axis-aligned annotation of the target dataset 1. To achieve this, we leverage the fact that neural networks are capable of learning richer representation of the target domain than what is utilized by the task. The under-utilized representation can be exploited to address a more detailed task. Our framework combines task knowledge of an out-of-domain source dataset with stronger annotation and domain knowledge of the target dataset with weaker annotation. A novel assignment process and projection loss are used to enable the co-training on the source and target datasets. As a result, the model is able to solve the more detailed task in the target domain, without additional computation overhead during inference. We extensively evaluate the method on various target datasets including fresh-produce dataset, HRSC2016 and SSDD. Results show that the proposed method consistently performs on par with the fully supervised approach.