Textual Inversion for Efficient Adaptation of Open-Vocabulary Object Detectors Without Forgetting

• URL: http://arxiv.org/abs/2508.05323v1
• Date: Thu, 07 Aug 2025 12:28:08 GMT
• Title: Textual Inversion for Efficient Adaptation of Open-Vocabulary Object Detectors Without Forgetting
• Authors: Frank Ruis, Gertjan Burghouts, Hugo Kuijf,
• Abstract summary: Textual Inversion (TI) allows extending the VLM vocabulary by learning new or improving existing tokens to accurately detect novel or fine-grained objects from as little as three examples.<n>The storage and gradient calculations are limited to the token embedding dimension, requiring significantly less compute than full-model fine-tuning.<n>We evaluate whether the method matches or outperforms the baseline methods that suffer from forgetting in a variety of quantitative and qualitative experiments.
• Score: 1.1871535995163365
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent progress in large pre-trained vision language models (VLMs) has reached state-of-the-art performance on several object detection benchmarks and boasts strong zero-shot capabilities, but for optimal performance on specific targets some form of finetuning is still necessary. While the initial VLM weights allow for great few-shot transfer learning, this usually involves the loss of the original natural language querying and zero-shot capabilities. Inspired by the success of Textual Inversion (TI) in personalizing text-to-image diffusion models, we propose a similar formulation for open-vocabulary object detection. TI allows extending the VLM vocabulary by learning new or improving existing tokens to accurately detect novel or fine-grained objects from as little as three examples. The learned tokens are completely compatible with the original VLM weights while keeping them frozen, retaining the original model's benchmark performance, and leveraging its existing capabilities such as zero-shot domain transfer (e.g., detecting a sketch of an object after training only on real photos). The storage and gradient calculations are limited to the token embedding dimension, requiring significantly less compute than full-model fine-tuning. We evaluated whether the method matches or outperforms the baseline methods that suffer from forgetting in a wide variety of quantitative and qualitative experiments.

Related papers

• Few-shot target-driven instance detection based on open-vocabulary object detection models [1.0749601922718608]
  Open-vocabulary object detection models bring closer visual and textual concepts in the same latent space. We propose a lightweight method to turn the latter into a one-shot or few-shot object recognition models without requiring textual descriptions.
  arXiv  Detail & Related papers  (2024-10-21T14:03:15Z)
• Black Box Few-Shot Adaptation for Vision-Language models [41.49584259596654]
  Vision-Language (V-L) models trained with contrastive learning to align the visual and language modalities have been shown to be strong few-shot learners. We describe a black-box method for V-L few-shot adaptation that operates on pre-computed image and text features. We propose Linear Feature Alignment (LFA), a simple linear approach for V-L re-alignment in the target domain.
  arXiv  Detail & Related papers  (2023-04-04T12:42:29Z)
• Scaling Novel Object Detection with Weakly Supervised Detection Transformers [21.219817483091166]
  We propose the Weakly Supervised Detection Transformer, which enables efficient knowledge transfer from a large-scale pretraining dataset to WSOD finetuning. Our experiments show that our approach outperforms previous state-of-the-art models on large-scale novel object detection datasets.
  arXiv  Detail & Related papers  (2022-07-11T21:45:54Z)
• Simple Open-Vocabulary Object Detection with Vision Transformers [51.57562920090721]
  We propose a strong recipe for transferring image-text models to open-vocabulary object detection. We use a standard Vision Transformer architecture with minimal modifications, contrastive image-text pre-training, and end-to-end detection fine-tuning. We provide the adaptation strategies and regularizations needed to attain very strong performance on zero-shot text-conditioned and one-shot image-conditioned object detection.
  arXiv  Detail & Related papers  (2022-05-12T17:20:36Z)
• Incremental-DETR: Incremental Few-Shot Object Detection via Self-Supervised Learning [60.64535309016623]
  We propose the Incremental-DETR that does incremental few-shot object detection via fine-tuning and self-supervised learning on the DETR object detector. To alleviate severe over-fitting with few novel class data, we first fine-tune the class-specific components of DETR with self-supervision. We further introduce a incremental few-shot fine-tuning strategy with knowledge distillation on the class-specific components of DETR to encourage the network in detecting novel classes without catastrophic forgetting.
  arXiv  Detail & Related papers  (2022-05-09T05:08:08Z)
• Multi-Modal Few-Shot Object Detection with Meta-Learning-Based Cross-Modal Prompting [77.69172089359606]
  We study multi-modal few-shot object detection (FSOD) in this paper, using both few-shot visual examples and class semantic information for detection. Our approach is motivated by the high-level conceptual similarity of (metric-based) meta-learning and prompt-based learning. We comprehensively evaluate the proposed multi-modal FSOD models on multiple few-shot object detection benchmarks, achieving promising results.
  arXiv  Detail & Related papers  (2022-04-16T16:45:06Z)
• ZSD-YOLO: Zero-Shot YOLO Detection using Vision-Language KnowledgeDistillation [5.424015823818208]
  A dataset such as COCO is extensively annotated across many images but with a sparse number of categories and annotating all object classes across a diverse domain is expensive and challenging. We develop a Vision-Language distillation method that aligns both image and text embeddings from a zero-shot pre-trained model such as CLIP to a modified semantic prediction head from a one-stage detector like YOLOv5. During inference, our model can be adapted to detect any number of object classes without additional training.
  arXiv  Detail & Related papers  (2021-09-24T16:46:36Z)
• Few-shot Weakly-Supervised Object Detection via Directional Statistics [55.97230224399744]
  We propose a probabilistic multiple instance learning approach for few-shot Common Object Localization (COL) and few-shot Weakly Supervised Object Detection (WSOD) Our model simultaneously learns the distribution of the novel objects and localizes them via expectation-maximization steps. Our experiments show that the proposed method, despite being simple, outperforms strong baselines in few-shot COL and WSOD, as well as large-scale WSOD tasks.
  arXiv  Detail & Related papers  (2021-03-25T22:34:16Z)
• Unsupervised Vision-and-Language Pre-training Without Parallel Images and Captions [92.47566804182338]
  We investigate if a strong V&L representation model can be learned through unsupervised pre-training without image-caption corpora. In particular, we propose to conduct mask-and-predict'' pre-training on text-only and image-only corpora. We find that such a simple approach performance close to a model pre-trained with aligned data, on four English V&L benchmarks.
  arXiv  Detail & Related papers  (2020-10-24T08:17:54Z)
• One-Shot Object Detection without Fine-Tuning [62.39210447209698]
  We introduce a two-stage model consisting of a first stage Matching-FCOS network and a second stage Structure-Aware Relation Module. We also propose novel training strategies that effectively improve detection performance. Our method exceeds the state-of-the-art one-shot performance consistently on multiple datasets.
  arXiv  Detail & Related papers  (2020-05-08T01:59:23Z)
• Incremental Few-Shot Object Detection [96.02543873402813]
  OpeN-ended Centre nEt is a detector for incrementally learning to detect class objects with few examples. ONCE fully respects the incremental learning paradigm, with novel class registration requiring only a single forward pass of few-shot training samples.
  arXiv  Detail & Related papers  (2020-03-10T12:56:59Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.