Foundation Model Priors Enhance Object Focus in Feature Space for Source-Free Object Detection

TL;DR

This paper introduces FALCON-SFOD, a novel framework that enhances object focus in feature space for source-free object detection under domain shift by leveraging foundation model priors and noise-robust pseudo-labeling.

Contribution

It proposes a new method combining foundation model-based regularization and imbalance-aware pseudo-labeling to improve object detection without source data.

Findings

Achieves competitive results on SFOD benchmarks.

Effectively improves object-focused feature representations.

Enhances pseudo-label quality under domain shift.

Abstract

Current state-of-the-art approaches in Source-Free Object Detection (SFOD) typically rely on Mean-Teacher self-labeling. However, domain shift often reduces the detector's ability to maintain strong object-focused representations, causing high-confidence activations over background clutter. This weak object focus results in unreliable pseudo-labels from the detection head. While prior works mainly refine these pseudo-labels, they overlook the underlying need to strengthen the feature space itself. We propose FALCON-SFOD (Foundation-Aligned Learning with Clutter suppression and Noise robustness), a framework designed to enhance object-focused adaptation under domain shift. It consists of two complementary components. SPAR (Spatial Prior-Aware Regularization) leverages the generalization strength of vision foundation models to regularize the detector's feature space. Using class-agnostic binary masks derived from OV-SAM, SPAR promotes structured and foreground-focused activations by guiding the network toward object regions. IRPL (Imbalance-aware Noise Robust Pseudo-Labeling) complements SPAR by promoting balanced and noise-tolerant learning under severe foreground-background imbalance. Guided by a theoretical analysis that connects these designs to tighter localization and classification error bounds, FALCON-SFOD achieves competitive performance across SFOD benchmarks.