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

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: VCR, Sairam, Lalla, Rishabh, Dayal, Aveen, Kulkarni, Tejal, Lalla, Anuj, Balasubramanian, Vineeth N, Khan, Muhammad Haris
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914340985110528
author VCR, Sairam
Lalla, Rishabh
Dayal, Aveen
Kulkarni, Tejal
Lalla, Anuj
Balasubramanian, Vineeth N
Khan, Muhammad Haris
author_facet VCR, Sairam
Lalla, Rishabh
Dayal, Aveen
Kulkarni, Tejal
Lalla, Anuj
Balasubramanian, Vineeth N
Khan, Muhammad Haris
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2512_17514
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Foundation Model Priors Enhance Object Focus in Feature Space for Source-Free Object Detection
VCR, Sairam
Lalla, Rishabh
Dayal, Aveen
Kulkarni, Tejal
Lalla, Anuj
Balasubramanian, Vineeth N
Khan, Muhammad Haris
Computer Vision and Pattern Recognition
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.
title Foundation Model Priors Enhance Object Focus in Feature Space for Source-Free Object Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.17514