HOMEY: Heuristic Object Masking with Enhanced YOLO for Property Insurance Risk Detection

Teerapong Panboonyuen

cs.CVMar 19, 2026

HOMEY: Heuristic Object Masking with Enhanced YOLO for Property Insurance Risk Detection

Authors: Teerapong Panboonyuen

Organizations: MARSAIL

Abstract

Automated property risk detection is a high-impact yet underexplored frontier in computer vision with direct implications for real estate, underwriting, and insurance operations. We introduce HOMEY (Heuristic Object Masking with Enhanced YOLO), a novel detection framework that combines YOLO with a domain-specific masking mechanism and a custom-designed loss function. HOMEY is trained to detect 17 risk-related property classes, including structural damages (e.g., cracked foundations, roof issues), maintenance neglect (e.g., dead yards, overgrown bushes), and liability hazards (e.g., falling gutters, garbage, hazard signs). Our approach introduces heuristic object masking to amplify weak signals in cluttered backgrounds and risk-aware loss calibration to balance class skew and severity weighting. Experiments on real-world property imagery demonstrate that HOMEY achieves superior detection accuracy and reliability compared to baseline YOLO models, while retaining fast inference. Beyond detection, HOMEY enables interpretable and cost-efficient risk analysis, laying the foundation for scalable AI-driven property insurance workflows.

Figures & tables

Figure 2: Sample property detection results on two real-world images. Each row shows: (1) original input image, (2) ground-truth labels, (3) baseline YOLO detection, and (4) our proposed HOMEY predictions. HOMEY demonstrates superior localization and classification of property risk elements, particularly in cluttered scenes.

Figure 3: Additional property detection examples showcasing the robustness of HOMEY. As in Figure 3 , each row contains original image, ground truth, baseline detection, and HOMEY output. Notice how HOMEY consistently captures subtle damages, overgrowth, and risk objects that the baseline model often misses.

Class	Instances	Precision		Recall		mAP 50-95
		Baseline	HOMEY	Baseline	HOMEY	Baseline	HOMEY
Bad Driveway	2	0.60	1.0	0.0	1.0	0.05	0.40
Boarded	172	0.71	0.77	0.50	0.55	0.42	0.62
Cracked Foundation	12	0.55	0.61	0.15	0.17	0.10	0.25
Damage	29	0.02	0.05	0.0	0.05	0.01	0.18
Dead Yard	6	0.23	0.26	0.0	0.12	0.05	0.22

Table 1: Quantitative comparison of HOMEY vs baseline across 17 property damage classes. HOMEY consistently outperforms baseline. Bold indicates best performance; color intensity scales with value for visual emphasis.

Figure 4: Training dynamics of the baseline: total loss and bounding-box loss.

Figure 5: Training dynamics of HOMEY: total loss and bounding-box loss. HOMEY converges faster and achieves lower loss values across epochs.

Figure 6: Confusion matrix of the baseline model across 17 property damage classes.

Figure 7: Confusion matrix of HOMEY across 17 property damage classes. Reduced off-diagonal errors demonstrate improved inter-class discrimination.

Figure 8: Qualitative results for the first set of property samples. Each row corresponds to an individual property, showing Original, Ground Truth, Baseline, and HOMEY predictions. HOMEY demonstrates more precise localization, better boundary delineation, and superior detection of subtle damages such as small cracks and minor roof issues, outperforming the baseline in every instance.

Figure 9: Qualitative results for the second set of property samples. HOMEY predictions are consistently closer to ground truth labels than baseline outputs. The model is particularly effective in detecting partially obscured damages and cluttered yard regions, illustrating its robustness to complex real-world scenarios.

Figure 10: Third set of samples comparing HOMEY against the baseline. HOMEY not only captures large-scale structural damages but also maintains accuracy for fine-grained issues like paint-rust and overgrowth, which are often missed by the baseline. This highlights HOMEY’s multi-scale detection capability.

Figure 11: Fourth set of property samples. HOMEY demonstrates significant improvements over baseline in scenarios with complex visual clutter, including overlapping vegetation and partially hidden objects. The predicted masks show higher fidelity and alignment with the ground truth, supporting more reliable property inspection.

Figure 12: Fifth set of property samples. HOMEY successfully identifies subtle and low-contrast damages such as cracked foundations and minor roof defects, maintaining consistent accuracy across different property types.

	YOLO12			YOLO26			HOMEY (Ours)
Class	P	R	mAP	P	R	mAP	P	R	mAP
Bad Driveway	0.58	0.00	0.04	0.62	0.10	0.08	1.00	1.00	0.40
Boarded	0.69	0.48	0.40	0.73	0.52	0.55	0.77	0.55	0.62
Cracked Foundation	0.52	0.12	0.08	0.58	0.14	0.18	0.61	0.17	0.25
Damage	0.01	0.00	0.01	0.03	0.02	0.10	0.05	0.05	0.18
Dead Yard	0.20	0.00	0.04	0.24	0.05	0.12	0.26	0.12	0.22

Table 2: Comparison of HOMEY with state-of-the-art YOLO variants on property risk detection. We report Precision (P), Recall (R), and mAP 50-95 across 17 classes. HOMEY consistently outperforms all baselines , especially on rare and high-risk categories.

Model	mAP 50-95	Recall	Precision
YOLO12	0.20	0.25	0.44
YOLO26	0.31	0.30	0.50
HOMEY (Ours)	0.40	0.36	0.56

Table 3: Overall performance comparison. HOMEY achieves the best balance between accuracy and robustness, particularly improving rare-class recall and overall mAP.

Class	YOLO26	HOMEY	Gain
Bad Driveway	0.08	0.40	+0.32
Cracked Foundation	0.18	0.25	+0.07
Falling Gutters	0.18	0.30	+0.12
Damage	0.10	0.18	+0.08
Roof Issues	0.03	0.05	+0.02
Paint-Rust Issues	0.08	0.12	+0.04

Table 4: Performance on rare and high-risk categories. HOMEY significantly improves the detection of underrepresented yet critical classes, demonstrating the effectiveness of heuristic masking and risk-aware loss.

Appendix figures & tables1 asset

Supplementary material from the paper’s appendix.

Appendix

Instance Details	Value
vCPUs	4
Memory (GiB)	16
Memory per vCPU (GiB)	4
Physical Processor	Intel Xeon Family
Clock Speed (GHz)	2.5
CPU Architecture	x86_64

Appendix

Table 5: Compute resources used for training and evaluation of HOMEY.

Object Detection

Object detection locates and classifies objects in images or video, balancing localization accuracy, robustness, and inference speed.

See all 624 papers

Uncertainty-Aware Vision-based Risk Object Identification via Conformal Risk Tube Prediction
Mar 25, 2026Kai-Yu Fu, Yi-Ting ChenAutonomous Driving Perception Autonomous Driving
National Yang Ming Chiao Tung University
Intelligent Spatial Estimation for Fire Hazards in Engineering Sites: An Enhanced YOLOv8-Powered Proximity Analysis Framework
Mar 10, 2026Ammar K. AlMhdawi, Nonso Nnamoko, Alaa Mashan Ubaid
University of Greater Manchester, A676 Deane Rd, Bolton, BL3 5AB, , UK · Edge Hill University, St Helens Rd, Ormskirk, L39 4QP, , UK · University of Khorfakkan, Al Jaradia , Sharjah, 887W+2P2, , UAE
YOLO-AMC: An Improved YOLO Architecture with Attention Mechanisms for Building Crack Detection
Jun 11, 2026Ching-Yu Tsai, Chia-Min Lin, Chih-Hsiang Yang +2Object Detection Attention Mechanisms
Department of Electrical and Computer Engineering, Tamkang University, Taiwan

Mar 25, 2026J/K move · Enter open · S save

Uncertainty-Aware Vision-based Risk Object Identification via Conformal Risk Tube Prediction

Kai-Yu Fu, Yi-Ting Chen

National Yang Ming Chiao Tung University

Autonomous Driving Perception Autonomous Driving Conformal Risk Control Autonomous Driving Safety Evaluation Object Detection

Open paper AskPDF arXiv

We study object importance-based vision risk object identification (Vision-ROI), a key capability for hazard detection in intelligent driving systems. Existing approaches make deterministic decisions and ignore uncertainty, which could lead to safety-critical failures. Specifically, in ambiguous scenarios, fixed decision thresholds may cause premature or delayed risk detection and temporally unstable predictions, especially in complex scenes with multiple interacting risks. Despite these challenges, current methods lack a principled framework to model risk uncertainty jointly across space and time. We propose Conformal Risk Tube Prediction, a unified formulation that captures spatiotemporal risk uncertainty, provides coverage guarantees for true risks, and produces calibrated risk scores with uncertainty estimates. To conduct a systematic evaluation, we present a new dataset and metrics probing diverse scenario configurations with multi-risk coupling effects, which are not supported by existing datasets. We systematically analyze factors affecting uncertainty estimation, including scenario variations, per-risk category behavior, and perception error propagation. Our method delivers substantial improvements over prior approaches, enhancing vision-ROI robustness and downstream performance, such as reducing nuisance braking alerts. For more qualitative results, please visit our project webpage: https://hcis-lab.github.io/CRTP/

Abstract

Object Detection

Explore similar work

Uncertainty-Aware Vision-based Risk Object Identification via Conformal Risk Tube Prediction

Intelligent Spatial Estimation for Fire Hazards in Engineering Sites: An Enhanced YOLOv8-Powered Proximity Analysis Framework

YOLO-AMC: An Improved YOLO Architecture with Attention Mechanisms for Building Crack Detection