CiF is a large new civil infrastructure segmentation dataset that shows zero-shot foundation models and domain-supervised models plateau at roughly 25% mAP, establishing infrastructure inspection as an open challenge for current visual AI.
hub Mixed citations
YOLOv11: An Overview of the Key Architectural Enhancements
Mixed citation behavior. Most common role is background (40%).
77
Pith papers citing it
503
external citations · Pith
Background
40% of classified citations
abstract
This study presents an architectural analysis of YOLOv11, the latest iteration in the YOLO (You Only Look Once) series of object detection models. We examine the models architectural innovations, including the introduction of the C3k2 (Cross Stage Partial with kernel size 2) block, SPPF (Spatial Pyramid Pooling - Fast), and C2PSA (Convolutional block with Parallel Spatial Attention) components, which contribute in improving the models performance in several ways such as enhanced feature extraction. The paper explores YOLOv11's expanded capabilities across various computer vision tasks, including object detection, instance segmentation, pose estimation, and oriented object detection (OBB). We review the model's performance improvements in terms of mean Average Precision (mAP) and computational efficiency compared to its predecessors, with a focus on the trade-off between parameter count and accuracy. Additionally, the study discusses YOLOv11's versatility across different model sizes, from nano to extra-large, catering to diverse application needs from edge devices to high-performance computing environments. Our research provides insights into YOLOv11's position within the broader landscape of object detection and its potential impact on real-time computer vision applications.
hub tools
citation-role summary
background 4
baseline 3
method 2
other 1
citation-polarity summary
representative citing papers
ScreenParse dataset and ScreenVLM model deliver dense screen parsing that outperforms larger VLMs on PageIoU and transfers to better UI grounding.
CineOrchestra unifies control of subjects, events, cameras, and shot transitions in cinematic video generation through entity-centric conditioning primitives and parameter-free coordinated rotary embeddings.
Future-L1 interleaves latent visual spans with text in MLLM decoding, trained on a custom Future-L1-50K dataset via LA-DAPO RL, and reports SOTA gains on FutureBench (61.0 to 85.4) and TwiFF-Bench (2.44 to 3.04).
A task-specific iterative framework for weakly supervised 4D radar scene flow estimation uses instance-aware self-supervised losses from 2D tracking/segmentation and a rigid static loss from odometry to outperform LiDAR-dependent cross-modal and fully supervised methods on the VoD dataset.
PoseBridge recovers semantic information lost during skeletonization by extracting pose-anchored cues from human pose estimation and transferring them via skeleton-conditioned bridging and semantic prototype adaptation, yielding 13.3-17.4 point gains on the Kinetics PURLS benchmark.
DPOFusion uses direct preference optimization on property-aligned and preference-controllable latent diffusion models to produce adaptive infrared-visible image fusions aligned with heterogeneous human and machine vision demands.
FluxShard uses per-block motion vectors and a Receptive Field Alignment Principle to manage feature cache reuse in edge-cloud video analytics, delivering 32.6-83.8% lower latency and 14.9-64.0% lower energy than baselines while preserving accuracy.
HELP uses heatmap-guided positional embeddings and a gradient mask to suppress background noise in queries, enabling efficient small-object detection with fewer decoder layers and parameters.
S2-CoT coordinates a Structural Fidelity Adapter in the encoder-decoder with a Semantic Context Adapter in the entropy model to convert potential performance loss into state-of-the-art gains across base codecs while using only a small fraction of parameters.
GeoMMBench reveals deficiencies in current multimodal LLMs for geoscience tasks while GeoMMAgent demonstrates that tool-integrated agents achieve significantly higher performance.
SARES-DEIM achieves 76.4% mAP50:95 and 93.8% mAP50 on HRSID by routing SAR features through sparse frequency and wavelet experts plus a high-resolution preservation neck, outperforming prior YOLO and SAR detectors.
UniSpector organizes visual prompt space with spatial-spectral and contrastive encoders to support open-set defect localization, beating baselines by at least 19.7% AP50b and 15.8% AP50m on the new Inspect Anything benchmark.
Gen-n-Val uses LLM and VLLM agents with Layer Diffusion and TextGrad to generate and validate synthetic instance data, cutting invalid samples from 50% to 7% and improving rare-class performance on LVIS and COCO benchmarks.
Orak is a foundational benchmark providing training data, interfaces, and evaluation tools for LLM agents across diverse video game genres.
A binary neural network fusing RAW frames and event streams achieves competitive low-light video enhancement at 20x lower compute than full-precision models.
SINA converts circuit schematic images to netlists at 96.67% accuracy using deep learning, OCR, connected-component labeling, and a vision-language model, claimed 2.72x better than prior methods.
RT-SFOD adapts dual-head detectors like YOLOv10 for source-free object detection via DHF pseudo-label fusion and MARD loss, delivering 1.4-3.5% mAP gains with 1.3x higher throughput and ~2x fewer parameters than prior SFOD methods.
Diffusion-generated, distribution-matched synthetic images enable zero-shot quantized object detectors to outperform prior zero-shot methods and even real-data QAT at 4-bit and 3-bit precision.
Proposes DERNet with Decompose-Enhance-Reconstruct operator and three plug-and-play modules to shift small object detection from spatial to spectral feature processing, claiming better performance than YOLOv11 with 1/6 the parameters.
BiliVLA combines supervised fine-tuning and GRPO on a 10k-frame phantom dataset to make a Qwen3-VL policy output category, bounding box, and 3-DoF commands, reporting 84.85% phantom success.
Boundary-by-Mask uses a foundation-model encoder plus SDF head to predict boundary-aware distance maps from few mask examples, enabling instance segmentation on low-texture industrial objects via SDF-to-mask reconstruction.
MambaCount uses S^4D blocks with spatial token selection and multi-granularity prototypes to reach MAE 12.23 on FSC-147 for open-vocabulary counting while keeping linear complexity.
Presents MMIOC-1M benchmark with 1M+ samples across 14 super-categories and RTVPNet with domain projection, sparse sampling, and bidirectional interaction, claiming SOTA on MMIOC-1M, LVIS, and COCO.
citing papers explorer
-
Cracks in the Foundation: A Civil Infrastructure Dataset to Challenge Vision Foundation Models
CiF is a large new civil infrastructure segmentation dataset that shows zero-shot foundation models and domain-supervised models plateau at roughly 25% mAP, establishing infrastructure inspection as an open challenge for current visual AI.
-
ScreenParse: Moving Beyond Sparse Grounding with Complete Screen Parsing Supervision
ScreenParse dataset and ScreenVLM model deliver dense screen parsing that outperforms larger VLMs on PageIoU and transfers to better UI grounding.
-
CineOrchestra: Unified Entity-Centric Conditioning for Cinematic Video Generation
CineOrchestra unifies control of subjects, events, cameras, and shot transitions in cinematic video generation through entity-centric conditioning primitives and parameter-free coordinated rotary embeddings.
-
Imagine Before You Predict: Interleaved Latent Visual Reasoning for Video Event Prediction
Future-L1 interleaves latent visual spans with text in MLLM decoding, trained on a custom Future-L1-50K dataset via LA-DAPO RL, and reports SOTA gains on FutureBench (61.0 to 85.4) and TwiFF-Bench (2.44 to 3.04).
-
Weakly Supervised Cross-Modal Learning for 4D Radar Scene Flow Estimation
A task-specific iterative framework for weakly supervised 4D radar scene flow estimation uses instance-aware self-supervised losses from 2D tracking/segmentation and a rigid static loss from odometry to outperform LiDAR-dependent cross-modal and fully supervised methods on the VoD dataset.
-
PoseBridge: Bridging the Skeletonization Gap for Zero-Shot Skeleton-Based Action Recognition
PoseBridge recovers semantic information lost during skeletonization by extracting pose-anchored cues from human pose estimation and transferring them via skeleton-conditioned bridging and semantic prototype adaptation, yielding 13.3-17.4 point gains on the Kinetics PURLS benchmark.
-
Fusion in Your Way: Aligning Image Fusion with Heterogeneous Demands via Direct Preference Optimization
DPOFusion uses direct preference optimization on property-aligned and preference-controllable latent diffusion models to produce adaptive infrared-visible image fusions aligned with heterogeneous human and machine vision demands.
-
FluxShard: Motion-Aware Feature Cache Reuse for Collaborative Video Analytics in Mobile Edge Computing
FluxShard uses per-block motion vectors and a Receptive Field Alignment Principle to manage feature cache reuse in edge-cloud video analytics, delivering 32.6-83.8% lower latency and 14.9-64.0% lower energy than baselines while preserving accuracy.
-
Learning Where to Embed: Noise-Aware Positional Embedding for Query Retrieval in Small-Object Detection
HELP uses heatmap-guided positional embeddings and a gradient mask to suppress background noise in queries, enabling efficient small-object detection with fewer decoder layers and parameters.
-
What and Where to Adapt: Structure-Semantics Co-Tuning for Machine Vision Compression via Synergistic Adapters
S2-CoT coordinates a Structural Fidelity Adapter in the encoder-decoder with a Semantic Context Adapter in the entropy model to convert potential performance loss into state-of-the-art gains across base codecs while using only a small fraction of parameters.
-
GeoMMBench and GeoMMAgent: Toward Expert-Level Multimodal Intelligence in Geoscience and Remote Sensing
GeoMMBench reveals deficiencies in current multimodal LLMs for geoscience tasks while GeoMMAgent demonstrates that tool-integrated agents achieve significantly higher performance.
-
SARES-DEIM: Sparse Mixture-of-Experts Meets DETR for Robust SAR Ship Detection
SARES-DEIM achieves 76.4% mAP50:95 and 93.8% mAP50 on HRSID by routing SAR features through sparse frequency and wavelet experts plus a high-resolution preservation neck, outperforming prior YOLO and SAR detectors.
-
UniSpector: Towards Universal Open-set Defect Recognition via Spectral-Contrastive Visual Prompting
UniSpector organizes visual prompt space with spatial-spectral and contrastive encoders to support open-set defect localization, beating baselines by at least 19.7% AP50b and 15.8% AP50m on the new Inspect Anything benchmark.
-
Gen-n-Val: Agentic Image Data Generation and Validation
Gen-n-Val uses LLM and VLLM agents with Layer Diffusion and TextGrad to generate and validate synthetic instance data, cutting invalid samples from 50% to 7% and improving rare-class performance on LVIS and COCO benchmarks.
-
Orak: A Foundational Benchmark for Training and Evaluating LLM Agents on Diverse Video Games
Orak is a foundational benchmark providing training data, interfaces, and evaluation tools for LLM agents across diverse video game genres.
-
EeveeDark: A Binary Neural Framework for Low-Light Video Enhancement via Event-Guided Sensor-Level Fusion
A binary neural network fusing RAW frames and event streams achieves competitive low-light video enhancement at 20x lower compute than full-precision models.
-
SINA: A Fully Automated Circuit Schematic Image to Netlist Generator Using Artificial Intelligence
SINA converts circuit schematic images to netlists at 96.67% accuracy using deep learning, OCR, connected-component labeling, and a vision-language model, claimed 2.72x better than prior methods.
-
Real-Time Source-Free Object Detection
RT-SFOD adapts dual-head detectors like YOLOv10 for source-free object detection via DHF pseudo-label fusion and MARD loss, delivering 1.4-3.5% mAP gains with 1.3x higher throughput and ~2x fewer parameters than prior SFOD methods.
-
Zero-Shot Quantization for Object Detectors using Off-the-Shelf Generative Models
Diffusion-generated, distribution-matched synthetic images enable zero-shot quantized object detectors to outperform prior zero-shot methods and even real-data QAT at 4-bit and 3-bit precision.
-
From Spatial to Spectral: An Efficient, Frequency-Guided Feature Representation Learner for Small Object Detection
Proposes DERNet with Decompose-Enhance-Reconstruct operator and three plug-and-play modules to shift small object detection from spatial to spectral feature processing, claiming better performance than YOLOv11 with 1/6 the parameters.
-
BiliVLA: Scene-Aware Vision-Language-Action Model with Reinforcement Learning for Autonomous Biliary Endoscopic Navigation
BiliVLA combines supervised fine-tuning and GRPO on a 10k-frame phantom dataset to make a Qwen3-VL policy output category, bounding box, and 3-DoF commands, reporting 84.85% phantom success.
-
Boundary-by-Mask: Few-Shot Instance Segmentation with Mask-Conditioned Boundary Learning for Texture-Poor Industrial Parts
Boundary-by-Mask uses a foundation-model encoder plus SDF head to predict boundary-aware distance maps from few mask examples, enabling instance segmentation on low-texture industrial objects via SDF-to-mask reconstruction.
-
MambaCount: Efficient Text-guided Open-vocabulary Object Counting with Spatial Sparse State Space Duality Block
MambaCount uses S^4D blocks with spatial token selection and multi-granularity prototypes to reach MAE 12.23 on FSC-147 for open-vocabulary counting while keeping linear complexity.
-
Unification of Closed-Open Industrial Detection Scenarios: New Large-Scale Benchmarks,Challenges and Baselines
Presents MMIOC-1M benchmark with 1M+ samples across 14 super-categories and RTVPNet with domain projection, sparse sampling, and bidirectional interaction, claiming SOTA on MMIOC-1M, LVIS, and COCO.
-
Policy-based Foveated Imaging and Perception
A task-aware policy learned via reinforcement learning allocates high-resolution pixels on dual-stream sensors in real time, outperforming fixed or non-predictive baselines under tight pixel budgets in both simulation and 200 MP hardware tests.
-
RefDiffNet: Learning to Expose Subtle PCB Defects Before Detection
RefDiffNet is a lightweight input enhancement block that uses reference image comparison to expose PCB defects, delivering up to 18% relative mAP50:95 gains across YOLO, RT-DETR, and Faster R-CNN detectors with 0.004-0.005M extra parameters.
-
AnyMo: Scaling Any-Modality Conditional Motion Generation with Masked Modeling
AnyMo is a masked-modeling framework for any-modality human motion generation trained on the new OmniHuMo dataset of 5,000+ hours of multimodal motion sequences.
-
EvalVerse: Pipeline-Aware and Expert-Calibrated Benchmarking for Professional Cinematic Video Generation
EvalVerse is a pipeline-aware benchmark that distills expert cinematic judgments into VLMs to assess 'goodness' metrics like aesthetics and multi-shot coherence alongside basic prompt adherence.
-
FedADAS: Communication-Efficient Federated Distillation for On-Device Driver Yawn Recognition in Vehicular Networks
FedADAS uses federated distillation to support heterogeneous on-device yawn recognition models across vehicles, delivering up to 9974x lower communication cost than standard federated learning while preserving accuracy under extreme data heterogeneity.
-
Contour-Native Bridge Defect Detection and Compact Digital Archiving with Frequency-Supervised Fourier Contours
FS-FSD regresses frequency-supervised Fourier contours for bridge defects, yielding higher polygon accuracy and better geometric quality than box, mask, or contour baselines on 3,767 UAV images with 42,346 instances.
-
Look Once, Beam Twice: Camera-Primed Real-Time Double-Directional mmWave Beam Management for Vehicular Connectivity
VIBE is a camera-primed hybrid model-based closed-loop learning system for real-time double-directional mmWave beam management in vehicular networks that achieves outage rates as low as 1.1-1.4% and outperforms 5G NR and end-to-end ML baselines.
-
Physical Adversarial Clothing Evades Visible-Thermal Detectors via Non-Overlapping RGB-T Pattern
Non-overlapping RGB-T adversarial patterns on clothing, optimized with spatial discrete-continuous optimization, achieve high attack success rates against multiple RGB-T detector fusion architectures in both digital and physical evaluations.
-
LIVEditor-14B: Lightning Unified Video Editing via In-Context Sparse Attention
LIVEditor-14B applies a new sparse attention method (ISA) that prunes context and uses query-sharpness routing to cut attention latency ~60% with no loss in editing quality on standard benchmarks.
-
Training-Free Tunnel Defect Inspection and Engineering Interpretation via Visual Recalibration and Entity Reconstruction
TunnelMIND recalibrates language-guided defect proposals via dense visual consistency and reconstructs them into structured defect entities with attributes for severity grading and retrieval-grounded engineering reports, reporting F1 scores of 0.68, 0.78, and 0.72 on visible, GPR, and road defect任务.
-
Transferable Physical-World Adversarial Patches Against Object Detection in Autonomous Driving
AdvAD produces physical-world adversarial patches with improved transferability to unseen object detectors by multi-model optimization, adaptive balancing, and physical variation robustness.
-
ZoomSpec: A Physics-Guided Coarse-to-Fine Framework for Wideband Spectrum Sensing
ZoomSpec achieves 78.1 mAP@0.5:0.95 on the SpaceNet dataset by combining log-space STFT, a coarse proposal net, adaptive heterodyne filtering, and dual-domain fine recognition to improve narrowband visibility in wideband spectrum sensing.
-
Toward Unified Fine-Grained Vehicle Classification and Automatic License Plate Recognition
UFPR-VeSV is a new real-world dataset for fine-grained vehicle classification and automatic license plate recognition collected from Brazilian police cameras, with benchmarks demonstrating its difficulty and the value of joint task use.
-
Visual Prototype Conditioned Focal Region Generation for UAV-Based Object Detection
UAVGen generates higher-quality synthetic UAV images via visual prototype conditioning and focal region focus in diffusion models, leading to better object detection accuracy than prior methods.
-
Chasing Ghosts: A Simulation-to-Real Olfactory Navigation Stack with Optional Vision Augmentation
A simulation-to-real navigation policy enables a quadrotor to locate an odor source using only basic olfaction sensors and optional vision, validated in indoor real-world flights.
-
VTC: DNN Compilation with Virtual Tensors for Data Movement Elimination
VTC eliminates data-movement operators in DNN graphs by representing intermediate tensors as index mappings, yielding up to 1.93x faster inference and 60% memory savings on GPUs.
-
PEPR: Privileged Event-based Predictive Regularization for Domain Generalization
PEPR reframes learning with privileged event data as predicting latent event features from RGB to improve domain generalization in object detection and segmentation without direct cross-modal alignment.
-
Population-Scale Advancing Interface Modeling Reveals How Bacterial Swarms Encode Future Spatial Architecture
A segmentation-plus-forecasting pipeline predicts the future shape of swarming bacterial colonies from their current boundary geometry, reaching 95.42% mIoU under an 80% observation / 20% prediction split.
-
Clutter-Robust Vision-Language-Action Models through Object-Centric and Geometry Grounding
OBEYED-VLA improves VLA robustness in cluttered real-world manipulation by disentangling perception into VLM-based object-centric grounding and geometry-aware stages, then fine-tuning the policy only on single-object demonstrations.
-
Edge Assisted Multi-Camera Vehicle Tracking Framework for Real-Time and Scalable Deployment
EASE-MCVT is a distributed edge-assisted multi-camera vehicle tracking framework that achieves real-time performance and competitive accuracy on public datasets through edge processing and server-side optimizations.
-
Video models are zero-shot learners and reasoners
Generative video models exhibit emergent zero-shot capabilities across perception, manipulation, and basic reasoning tasks.
-
Synthetic Data Augmentation for Enhanced Chicken Carcass Instance Segmentation
Synthetic data augmentation improves instance segmentation performance for chicken carcasses when real annotated data is limited.
-
A Leaf-Level Dataset for Soybean-Cotton Detection and Segmentation
A new leaf-instance dataset for soybean-cotton detection and segmentation collected across growth stages and conditions from commercial farms is presented and validated with YOLOv11.
-
FSDC-DETR: A Frequency-Spatial Domain Collaborative DETR for Small Object Detection
FSDC-DETR improves small-object AP by 6.8–6.9 points on VisDrone and AITODv2 by explicit frequency-spatial fusion and wavelet-style downsampling inside a DETR hybrid encoder.
-
Autonomous UAV Navigation for Individual Wildlife Re-Identification
Autonomous UAV navigation system integrates YOLOv11 and DINOv2 to optimize real-time flight for capturing re-ID quality images of patterned wildlife, demonstrated feasible on Kenyan zebras with claimed generalization to other species.
-
Hippocampus-DETR: An Explicit Memory Object Detection Framework Based on Hippocampus Modeling
Hippocampus-DETR integrates a hippocampal memory network (HipNet) into DETR to simulate brain subregions for pattern separation, completion, and improved detection accuracy plus generalization.