# PISA: A Pseudo-Individual Source-Domain Feature Adaptation Framework for Test-Time Open-Vocabulary Object Detection

arXiv: https://arxiv.org/abs/2608.14142 · HTML: https://mlsanity.com/paper/2608.14142/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 31 references that are in the map.
- First posted on arXiv 14 Aug 2026.

## Builds on (most cited first, 18 of 18)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://mlsanity.com/paper/2303.05499/) (2023)
- [Benchmarking Neural Network Robustness to Common Corruptions and Perturbations](https://mlsanity.com/paper/1903.12261/) (2019)
- [LVIS: A Dataset for Large Vocabulary Instance Segmentation](https://mlsanity.com/paper/1908.03195/) (2019)
- [A Comprehensive Survey on Test-Time Adaptation under Distribution Shifts](https://mlsanity.com/paper/2303.15361/) (2023)
- [Benchmarking Robustness in Object Detection: Autonomous Driving when Winter is Coming](https://mlsanity.com/paper/1907.07484/) (2019)
- [Cross-Domain Weakly-Supervised Object Detection through Progressive Domain Adaptation](https://mlsanity.com/paper/1803.11365/) (2018)
- [Generalized Source-free Domain Adaptation](https://mlsanity.com/paper/2108.01614/) (2021)
- [Towards Open Vocabulary Learning: A Survey](https://mlsanity.com/paper/2306.15880/) (2023)
- [Source-Free Domain Adaptation via Distribution Estimation](https://mlsanity.com/paper/2204.11257/) (2022)
- [VDM-DA: Virtual Domain Modeling for Source Data-free Domain Adaptation](https://mlsanity.com/paper/2103.14357/) (2021)
- [5%>100%: Breaking Performance Shackles of Full Fine-Tuning on Visual Recognition Tasks](https://mlsanity.com/paper/2408.08345/) (2024)
- [Transfer CLIP for Generalizable Image Denoising](https://mlsanity.com/paper/2403.15132/) (2024)
- [What, How, and When Should Object Detectors Update in Continually Changing Test Domains?](https://mlsanity.com/paper/2312.08875/) (2023)
- [Open-Vocabulary Object Detectors: Robustness Challenges under Distribution Shifts](https://mlsanity.com/paper/2405.14874/) (2024)
- [VLOD-TTA: Test-Time Adaptation of Vision-Language Object Detectors](https://mlsanity.com/paper/2510.00458/) (2025)
- [Test-time Vocabulary Adaptation for Language-driven Object Detection](https://mlsanity.com/paper/2506.00333/) (2025)

## Abstract (arXiv)

Open-vocabulary object detection test-time adaptation (OVOD-TTA) aims to address the performance degradation that pre-trained base models suffer when encountering image-domain shifts. Existing source-free OVOD-TTA methods rely either on refined test-time information for re-scoring or on pseudo-labels for self-training, leading to significant accuracy degradation when initial predictions are poor. Meanwhile, most conventional source-domain estimation methods recover abstract, sparse representations suitable for the classification task, but fail to capture the dense, concrete features required for detection. To address these issues, we propose PISA, a novel source-free OVOD-TTA method that can be seamlessly integrated into open-vocabulary visual backbones. The core components of our method are the Corruption-Invariant Feature Extractor (CIFE), the Feature Alignment Module (FAM), and a multi-scale alignment framework (BAA). To capture detection-suitable features, we develop CIFE to exploit the invariance of CLIP's visual features across corrupted images, ensuring robustness against various corruptions. We further develop FAM and BAA for the pre-training and adaptation to transform the corruption-invariant features into pseudo-individual source-domain features that are close to the original source-domain features. In this way, dense and concrete pseudo-individual source-domain features are used for supervision instead of unreliable pseudo-label signals. Experiments on the corrupted VOC-C, COCO-C, and LVIS-C benchmarks across three base models demonstrate that PISA substantially improves both the localization precision and the category recognition accuracy of the original models. Notably, PISA achieves state-of-the-art performance without requiring access to source-domain data, surpassing existing methods by 3.92% in AP@50% on COCO-C.