Computer Vision

Novel implicit field representation accurately reconstructs multi-person interaction geometry in 3D, simultaneously capturing occupancy, instance IDs, and contact fields, surpassing existing methods.

Researchers developed semantics-aware adversarial examples using a probabilistic approach, achieving higher success rates in bypassing defenses while remaining undetectable to humans.

Trust Sampling enhances guided diffusion by iteratively optimizing constrained generation at each step, improving efficiency and accuracy in image and 3D motion generation.

Consistency Purification boosts certified robustness by efficiently purifying noisy images using a one-step generative model, achieving state-of-the-art results while maintaining semantic alignment.

Consistency Diffusion Bridge Models (CDBMs) dramatically speed up diffusion bridge model sampling by learning a consistency function, achieving up to a 50x speedup with improved sample quality.

SeCo: Semantic Connectivity-driven Pseudo-Labeling enhances cross-domain semantic segmentation by correcting noisy pseudo-labels at the connectivity level, improving model accuracy and robustness.

Conditional Controllable Fusion (CCF) achieves training-free, adaptable image fusion by dynamically injecting fusion conditions into a pre-trained denoising diffusion model.

AutoPalette: a new framework minimizing color redundancy in dataset distillation, resulting in more efficient model training with comparable performance.

Collaborative Video Diffusion (CVD) generates multiple consistent videos from various camera angles using a novel cross-video synchronization module, significantly improving consistency compared to ex…

ColJailBreak cleverly circumvents AI safety filters by first generating safe images and then subtly injecting unsafe content using image editing.

Coherent 3D scenes are diffused from a single RGB image using a novel image-conditioned 3D scene diffusion model, surpassing state-of-the-art methods.

CoFie: A novel local geometry-aware neural surface representation dramatically improves accuracy and efficiency in 3D shape modeling by using coordinate fields to compress local shape information.

Hierarchical concept bottleneck models boost interpretability and accuracy in visual classification by uncovering both high-level and low-level concepts.

Researchers developed CNCA, a novel method that generates realistic and customizable adversarial camouflage for vehicle detectors by leveraging a pre-trained diffusion model, surpassing existing metho…

Cluster-Learngene efficiently initializes elastic-scale Vision Transformers by adaptively clustering and inheriting key modules from a large ancestry model, saving resources and boosting downstream ta…

COIN: A novel method for Cloud Object Detector Adaptation that integrates knowledge from cloud models and CLIP to train highly accurate target detectors, achieving state-of-the-art performance.

Classification Diffusion Models (CDMs) revolutionize density ratio estimation by integrating the strengths of diffusion models and classifiers, achieving state-of-the-art image generation and likeliho…

CHASE: A novel method for skeleton-based multi-entity action recognition that cleverly adapts skeleton positions to minimize data bias and boost accuracy.

CemiFace: Generating high-quality synthetic facial data for robust face recognition, while addressing privacy concerns.

Learned image compression gets a boost with a novel Causal Context Adjustment Loss, improving efficiency without sacrificing quality.