Train, Test, Re-evaluate: Schedule-Sensitive Evaluation of Generative Data for Hand Detection

• URL: http://arxiv.org/abs/2606.01896v1
• Date: Mon, 01 Jun 2026 08:38:25 GMT
• Title: Train, Test, Re-evaluate: Schedule-Sensitive Evaluation of Generative Data for Hand Detection
• Abstract summary: We train YOLOv8n hand detectors under six training-and-scheduling regimes.<n>Experiments extract substantial real-deployment benefit from inpainted accessory data.
• Score: 0.5161531917413708
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Generated (or synthetic) image data is increasingly used to augment or replace real training datasets when target imagery is scarce, expensive, or biased. For hand detection, particularly in occupational safety settings, public datasets mostly contain bare hands. This under-represents the variation in hand appearance introduced by gloves, tattoos, jewelry, and other personal protective equipment, creating a distribution shift that safety-critical applications encounter at deployment. We test whether generative inpainting, editing only the hand region of a real photograph to introduce accessories, can close this shift gap. On a paired dataset of real images and their synthetic counterparts, we train YOLOv8n hand detectors under six training-and-scheduling regimes (Experiments A-F, three random seeds each), evaluate every detector on a real test set and on a real-gloves-only test split, and report the mean average precision (mAP) at two overlap thresholds (mAP@0.5 and mAP@0.5:0.95) along with paired statistical tests. A two-stage experiment: train on real U synthetic data, then fine-tune the resulting weights on real-only at a lower learning rate, increases mAP@0.5 compared to the real-only baseline model on the standard real test set, and improves the real-gloves out-of-distribution gap. Another three-stage experiment preserves box-tightness best, reaching the highest mAP@0.5:0.95 of any other experiment in the study. The synthetic-data utility for safety-critical hand detection is determined by the training procedure, and simple multi-stage experiments extract substantial real-deployment benefit from inpainted accessory data.

Related papers

• Should I Use This Synthetic Dataset for Training? How to Test with Minimal Real Data [73.19814469893088]
  Digital twins and learned world models are increasingly used to generate synthetic data that augment the scarce real datasets available for training artificial intelligence (AI) models in engineering systems.<n>This paper addresses whether training on the augmented dataset improves the true, population-level performance, while consuming as few real test data points as possible.<n>Experiments on a synthetic-data classification task, a DT-aided wireless packet-scheduling task, and a radio-map prediction task show that aeSFT identifies useful synthetic data using substantially fewer real test samples than mean-based sequential testing.
  arXiv  Detail & Related papers  (2026-08-28T07:05:38Z)
• Image Augmentation as Test Generation for Deep Learning-Based Image Retrieval Systems [5.527117132133789]
  This paper reviews augmentation and generation techniques and evaluates them as test generators for embedding-based image retrieval systems.<n>Weather simulation and SaSPA are the image augmentation/generation techniques that produce the highest embedding uncertainty and failure rates.<n>GAN-based augmentation techniques are among the lowest in realism, indicating the presence of synthetic artifacts and perceptual inconsistencies.
  arXiv  Detail & Related papers  (2026-08-27T03:40:00Z)
• ScreenShot: A Foundation Model for Few-Shot Combination Drug Screening [2.971294564098473]
  ScreenShot is a hierarchical transformer pretrained on 40 drug screening datasets covering 3,700 drugs and 6,000 biological samples.<n>Given a few-shot context of observations from a new patient, ScreenShot predicts the response of the sample to combination therapies through in-context learning.<n>On four held-out datasets, ScreenShot outperforms all baselines in both prediction accuracy and identification of selectively effective treatments.
  arXiv  Detail & Related papers  (2026-08-12T16:13:50Z)
• Balancing Real and Synthetic Data for CNN-based Masonry Crack Detection [37.63495002959663]
  Cracks are a critical indicator of building health, and early stage identification is fundamental to prevent harmful damages.<n>To address this limitation, we explored generating synthetic crack data, which complements real data and improves training effectiveness.<n>The method demonstrates the potential of synthetic data to reduce collection efforts while enhancing crack detection accuracy.
  arXiv  Detail & Related papers  (2026-06-06T07:51:32Z)
• Few-Shot Synthetic Data Generation with Diffusion Models for Downstream Vision Tasks [36.94429692322632]
  Class imbalance is a persistent challenge in visual recognition.<n>We propose a lightweight synthetic data augmentation pipeline that fine-tunes a LoRA adapter on as few as 20-50 real images of a rare class.
  arXiv  Detail & Related papers  (2026-05-12T10:11:57Z)
• Synthetic Defect Image Generation for Power Line Insulator Inspection Using Multimodal Large Language Models [3.72557681438567]
  Utility companies increasingly rely on drone imagery for routine inspection, but training accurate defect-type classifiers remains difficult.<n>We address this data-scarcity setting by using an off-the-shelf multimodal large language model (MLLM) as a training-free image generator.<n>We evaluate on ceramic insulator defect-type classification using a public dataset with a realistic low training-data regime.
  arXiv  Detail & Related papers  (2026-03-09T08:06:27Z)
• RealStats: A Rigorous Real-Only Statistical Framework for Fake Image Detection [3.4104142774745996]
  We introduce a rigorous, statistically grounded framework for fake image detection that focuses on producing a probability score interpretable with respect to the real-image population.<n>This framework is generic, flexible, and training-free, making it well-suited for robust fake image detection across diverse and evolving settings.
  arXiv  Detail & Related papers  (2026-01-26T19:11:31Z)
• Synthetic-to-Real Object Detection using YOLOv11 and Domain Randomization Strategies [0.0]
  This paper focuses on training a YOLOv11 model to detect a specific object (a soup can) using only synthetic data and domain randomization strategies.<n>The methodology involves extensive experimentation with data augmentation, dataset composition, and model scaling.<n>The best performing configuration, a YOLOv11l model trained on an expanded and diverse dataset, achieved a final mAP@50 of 0.910 on the competition's hidden test set.
  arXiv  Detail & Related papers  (2025-09-18T15:08:54Z)
• Testing Human-Hand Segmentation on In-Distribution and Out-of-Distribution Data in Human-Robot Interactions Using a Deep Ensemble Model [40.815678328617686]
  We present a novel approach by evaluating the performance of pre-trained deep learning models under both ID data and more challenging OOD scenarios.<n>We incorporated unique and rare conditions such as finger-crossing gestures and motion blur from fast-moving hands.<n>Results revealed that models trained on industrial datasets outperformed those trained on non-industrial datasets.
  arXiv  Detail & Related papers  (2025-01-13T21:52:46Z)
• XLD: A Cross-Lane Dataset for Benchmarking Novel Driving View Synthesis [84.23233209017192]
  This paper presents a synthetic dataset for novel driving view synthesis evaluation.<n>It includes testing images captured by deviating from the training trajectory by $1-4$ meters.<n>We establish the first realistic benchmark for evaluating existing NVS approaches under front-only and multicamera settings.
  arXiv  Detail & Related papers  (2024-06-26T14:00:21Z)
• From Fake to Real: Pretraining on Balanced Synthetic Images to Prevent Spurious Correlations in Image Recognition [64.59093444558549]
  We propose a simple, easy-to-implement, two-step training pipeline that we call From Fake to Real. By training on real and synthetic data separately, FFR does not expose the model to the statistical differences between real and synthetic data. Our experiments show that FFR improves worst group accuracy over the state-of-the-art by up to 20% over three datasets.
  arXiv  Detail & Related papers  (2023-08-08T19:52:28Z)
• Seamless Iterative Semi-Supervised Correction of Imperfect Labels in Microscopy Images [57.42492501915773]
  In-vitro tests are an alternative to animal testing for the toxicity of medical devices. Human fatigue plays a role in error making, making the use of deep learning appealing. We propose Seamless Iterative Semi-Supervised correction of Imperfect labels (SISSI) Our method successfully provides an adaptive early learning correction technique for object detection.
  arXiv  Detail & Related papers  (2022-08-05T18:52:20Z)
• Inertial Hallucinations -- When Wearable Inertial Devices Start Seeing Things [82.15959827765325]
  We propose a novel approach to multimodal sensor fusion for Ambient Assisted Living (AAL) We address two major shortcomings of standard multimodal approaches, limited area coverage and reduced reliability. Our new framework fuses the concept of modality hallucination with triplet learning to train a model with different modalities to handle missing sensors at inference time.
  arXiv  Detail & Related papers  (2022-07-14T10:04:18Z)
• Leveraging Unlabeled Data to Predict Out-of-Distribution Performance [63.740181251997306]
  Real-world machine learning deployments are characterized by mismatches between the source (training) and target (test) distributions. In this work, we investigate methods for predicting the target domain accuracy using only labeled source data and unlabeled target data. We propose Average Thresholded Confidence (ATC), a practical method that learns a threshold on the model's confidence, predicting accuracy as the fraction of unlabeled examples.
  arXiv  Detail & Related papers  (2022-01-11T23:01:12Z)
• Federated Test-Time Adaptive Face Presentation Attack Detection with Dual-Phase Privacy Preservation [100.69458267888962]
  Face presentation attack detection (fPAD) plays a critical role in the modern face recognition pipeline. Due to legal and privacy issues, training data (real face images and spoof images) are not allowed to be directly shared between different data sources. We propose a Federated Test-Time Adaptive Face Presentation Attack Detection with Dual-Phase Privacy Preservation framework.
  arXiv  Detail & Related papers  (2021-10-25T02:51:05Z)

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.