Computer Vision

MF-VPD reduces model parameters by nearly 77% while boosting performance in visual perception tasks, making it a game-changer for efficient AI applications.

Small forward-marginal errors can mask significant numerical instability in diffusion sampling, leading to misleading conclusions about model performance.

BioModule transforms any 3D pose estimator into a tool for biomechanical analysis, bridging the gap between geometric accuracy and physical interpretability.

ProsMAE outperforms traditional methods by leveraging diverse histopathology datasets to significantly improve ISUP grade classification accuracy.

QR codes can now be secured against spoofing attacks with a dual-mode architecture that balances offline integrity and real-time validation.

SAM-MT decouples latency from target count, enabling real-time video segmentation that rivals single-target performance even with multiple objects in view.

PanoLOG's innovative partitioning strategy transforms panoramic 3D reconstruction, achieving top-tier rendering quality while slashing computational costs.

Event-based lip reading accuracy improves dramatically when leveraging viseme-aware temporal modeling, revealing the critical role of motion trajectories in distinguishing lip movements.

Fine-grained textual cues can dramatically improve face attack detection, revealing vulnerabilities in existing systems that rely solely on visual data.

UAV-OVVIS enables flexible target detection in UAV videos, outperforming traditional methods by allowing open-vocabulary queries for instance-level segmentation.

Canvas360 not only redefines panoramic generation with geometry-aware techniques but also delivers a dataset of 1 million samples that transforms how we approach in-context tasks.

By leveraging outdated DEMs, this framework achieves real-time, high-fidelity 3D terrain reconstruction, transforming wildfire hazard assessments.

A novel AI-driven approach reveals that targeted facial stimuli can significantly enhance the detection of perceptual differences between autistic and neurotypical individuals.

Surpassing larger models, this agent achieves 91.4% retrieval accuracy in long-horizon multimodal dialogues by leveraging episodic memory for efficient context management.

An ensemble approach combining U-Net and a geospatial model can accurately predict viticulture potential, outperforming traditional assessment methods.

A single DNN model can dynamically adjust input resolution for LiDAR object detection, achieving superior performance and efficiency in real-time applications.

A unified benchmark that evaluates federated learning in medical imaging across multiple organs reveals critical gaps in existing assessments and emphasizes the need for efficiency and privacy metrics.

Structured distribution shifts can significantly impair TWR HAR performance, but a new theoretical framework reveals how to enhance generalization across varying conditions.

ConRad boosts the efficiency of radiomic feature predictions by integrating segmentation boundary uncertainty, offering a game-changing approach to clinical imaging reliability.

Information restriction is the key to understanding how diffusion models can generalize rather than memorize, revealing a precise phase boundary that could transform generative AI practices.

Vision-language models struggle with safety-critical reasoning, but AUTOPILOT-VQA reveals their limitations in understanding real-world driving incidents.

Track2Map achieves real-time 3D reconstruction in robotic surgery, even when camera trajectory data is unreliable or missing.

VocaDet enables open-vocabulary object detection that evolves with user input, achieving high performance without retraining the model.

Roop's face-swapping model achieves a breakthrough in balancing pedestrian privacy and data usability, outperforming existing methods.

MobiDiff achieves a 5.3x speedup in generating synthetic mobility data while maintaining high fidelity to real-world patterns.

A novel hybrid approach boosts rare-class instance segmentation performance by up to 9.5 AP points by intelligently combining T2I generation with context-aware I2I editing.

State-of-the-art performance in endoscopic referring segmentation is now achievable through a novel attribute retrieval approach, transforming how we interpret complex medical imagery.

Classical symmetry scoring methods can rival deep learning approaches in performance while being orders of magnitude faster, challenging the assumption that deeper networks always outperform traditional techniques.

LightCrafter achieves superior video relighting by integrating PBR with diffusion models, enabling intricate lighting control and long-form temporal consistency without the need for extensive training data.

Vision Transformers outperform CNNs in modeling human texture perception, revealing a fundamental shift in how we understand visual processing in AI.

ARGUS achieves up to 97% tracking accuracy in under a minute, revolutionizing automated cell tracking without the need for training data or GPU support.

Integrating geographical encoding with a robust data quality assessment reduces poverty prediction errors by nearly 19% in satellite imagery analysis.

Traditional generative models struggle with subtle neurodegenerative changes, but Latent Drift captures clinically relevant progression by focusing on compressed semantic representations.

Multimodal queries in UAV imagery can significantly reduce visual-query ambiguity and improve target localization across diverse scenarios.

Multi-deformation modeling can be achieved more effectively by choosing the right integration strategy, impacting the robustness of dynamic scene reconstruction.

HSA achieves up to 41.5% improvement in scene decomposition accuracy by leveraging hierarchical semantic understanding with minimal labeled data.

Adapting pretrained models on-the-fly to ever-changing data distributions could redefine how we deploy AI in dynamic environments.

SkelGen4D achieves high-quality text-driven mesh animation without the burden of extensive skeleton annotations, outperforming fully supervised models.

StatLUT achieves artifact-free photorealistic style transfer by decoupling color and structure, outperforming state-of-the-art methods in visual fidelity.

GRE-Diff enables users to create and refine apartment layouts interactively, merging AI efficiency with human creativity in unprecedented ways.

LUMI revolutionizes lossless image compression by decoupling it from tokenizer behavior, achieving superior performance with a unified approach across different LLM architectures.

Performance claims in colonoscopy polyp segmentation may be misleading, with a single metric shift altering the perceived best model.

Achieving a superior balance between domain consistency and information preservation in joint distribution modeling could redefine approaches to unpaired data scenarios.

Achieving state-of-the-art accuracy in RGB-Thermal video object detection, DHNet tackles spatial misalignment with innovative dual-correlation learning.

EVIS bridges the gap between simulation and real-world event-camera data, enabling rapid prototyping and testing for robotics without the costly data collection process.

SAGA boosts temporal stability in autoregressive video generation, achieving a remarkable increase in temporal quality from 97.30 to 97.91 without retraining.

Textual prompts can revolutionize object detection in aerial imagery, enabling models to adaptively focus on complex scenes with unprecedented accuracy.

RadLoc achieves unprecedented speed and robustness in radar-based global localization, outperforming state-of-the-art methods while using the smallest descriptor size.

Traditional tetrahedralization is error-prone, but HoloTetSphere achieves a unified, coherent mesh that enhances physical simulation accuracy.

ZipDepth achieves real-time monocular depth estimation on resource-constrained devices while rivaling the accuracy of much larger models.