FlowOVD: Learning Generative Latent Flows for
Zero-shot Open-vocabulary Detection

Yao Wei1, Andrea Cavallaro2, Changjae Oh1
1Queen Mary University of London, UK
2École Polytechnique Fédérale de Lausanne (EPFL), Switzerland

arXiv 2026

arXiv Code (coming soon)

Traditionally, OVD is formulated as a discriminative prediction task through a Transformer encoder-decoder architecture. By introducing a generative perspective, our work attempts to move beyond conventional discrete query construction strategies.

Abstract

Open-vocabulary object detection (OVD) has achieved remarkable progress through large-scale vision-language pre-training. Existing methods, however, typically formulate OVD as a discriminative prediction problem, where decoder queries are either static or initialized from encoder features, thus limiting their diversity and flexibility. In this paper, we introduce a generative perspective by modeling decoder query generation as a continuous transport process in latent space. We propose FlowOVD, a text-conditioned query generation framework based on rectified flow that progressively transforms text-agnostic queries into text-guided queries. By introducing continuous latent query dynamics into a vision-language model (VLM) based detector, our method avoids heuristic discrete query construction and enables more expressive semantic alignment for open-vocabulary detection. Without requiring additional training data, FlowOVD achieves 49.5 AP on COCO and 31.5 AP on LVIS, outperforming GroundingDINO by +1.2 AP (+2.5 %) and +4.1 AP (+15.0 %), respectively. The larger gain on the challenging long-tailed LVIS benchmark further highlights the effectiveness of continuous query generation for open-vocabulary generalization.

Overview

We presented FlowOVD, a novel OVD approach that reformulates query initialization as a continuous generative process in the latent space. We propose a text-conditioned query flow that transforms an initial set of text-agnostic queries into a text-guided distribution. Specifically, we adopt a rectified flow formulation to model a time-dependent velocity field that transports queries under language conditioning. This enables controllable, and diverse query generation while remaining fully compatible with existing Transformer-based detectors.

(a) Overall framework. Given an image and a text prompt, a vision-language encoder extracts visual features and textual features. In the latent space, we present a query flow that transforms a set of text-agnostic queries into text-conditioned queries. The refined queries are then consumed by the transformer decoder to produce final predictions. (b) Query Flow. First, a set-level matching is performed between source queries and target queries. Intermediate states are obtained via linear interpolation. A velocity field is trained to learn the transport dynamics with conditions. The final queries are obtained by integrating the learned flow, resulting in more diverse and text-aligned query representations.

Result

Extensive experiments on COCO and LVIS show that our FlowOVD consistently improves open-vocabulary detection performance and achieves better efficiency with fewer decoder layers.

BibTeX

@article{wei2026flowovd,
      title={FlowOVD: Learning Generative Latent Flows for Zero-shot Open-vocabulary Detection},
      author={Wei, Yao and Cavallaro, Andrea and Oh, Changjae},
      journal={arXiv preprint arXiv:2606.00782},
      year={2026}}