Computer Vision

CYCLO: A novel cyclic graph transformer excels at multi-object relationship modeling in aerial videos.

CV-VAE: A compatible video VAE enabling efficient, high-quality latent video generation by bridging the gap between image and video latent spaces.

CUFIT: a novel curriculum fine-tuning paradigm significantly improves medical image classification accuracy despite noisy labels by leveraging pre-trained Vision Foundation Models.

Ctrl-X: Zero-shot text-to-image generation with training-free structure & appearance control!

CryoSPIN revolutionizes ab-initio cryo-EM reconstruction with semi-amortized pose inference, achieving faster and more accurate 3D structure determination.

CryoGEM: Physics-informed generative model creates realistic synthetic cryo-EM datasets, boosting particle picking and pose estimation accuracy for higher-resolution protein structure determination.

Cross-video Identity-cOrrelating pre-training (CION) revolutionizes person re-identification by leveraging identity correlation across videos for superior model pre-training, achieving state-of-the-ar…

Self-supervised blind image deblurring (BID) breakthrough! A novel cross-scale consistency loss and progressive training scheme using implicit neural representations achieves superior performance wit…

Cross-Modality Perturbation Synergy (CMPS) attack: A novel universal perturbation method for cross-modality person re-identification, effectively misleading ReID models by leveraging gradients from di…

CRAYM: Neural field optimization via camera RAY matching enhances 3D reconstruction by using camera rays, not pixels, improving both novel view synthesis and geometry.

COVE: Consistent high-quality video editing achieved by leveraging diffusion feature correspondence for temporal consistency.

COUNTGD: A new multi-modal model counts objects in images using text or visual examples, significantly improving open-world counting accuracy.

CoSW: a novel conditional sample weighting method for robust smoke segmentation, achieves state-of-the-art results by handling inconsistent noisy labels through a multi-prototype framework.

COSMIC efficiently compresses satellite images via a lightweight encoder and diffusion compensation, enabling practical onboard processing and high compression ratios.

CosAE: a novel autoencoder using learnable Fourier series achieves state-of-the-art image restoration by encoding frequency coefficients in its narrow bottleneck, preserving fine details even with ext…

Federated Hardware-Prompt Learning (FedHP) enables robust cross-hardware SCI training by aligning inconsistent data distributions using a hardware-conditioned prompter, outperforming existing FL metho…

Self-supervised learning models can now better predict primate IT neural responses by preserving structured variability to input transformations, improving alignment with biological visual perception.

Spiking neural network effectively segments mixed-motion event streams via spike-based Bayesian computation, achieving efficient real-time motion decoupling.

NerPE: continuous heatmap regression via implicit neural representation resolves the accuracy-limiting quantization errors in human pose estimation, achieving sub-pixel precision.

ContextGS: Revolutionizing 3D scene compression with an anchor-level autoregressive model, achieving 15x size reduction in 3D Gaussian Splatting while boosting rendering quality.