Computer Vision

OmnimatteZero: Real-time omnimatte using pre-trained video diffusion, no training needed!

RDTF: Efficient animated sticker generation via dual-mask training, outperforming parameter-efficient tuning under constrained resources.

4D-Bench: The first benchmark for assessing MLLMs in 4D object understanding, revealing weak temporal understanding and the need for advancements.

Adaptive Diffusion Models with Minority-Aware Adaptive DPO

TaoAvatar: Lifelike talking avatars in AR, using 3D Gaussian Splatting for real-time rendering and high fidelity.

Presents a strong baseline for multi-UAV tracking in thermal infrared video using YOLOv12 and BoT-SORT, achieving competitive results without complex enhancements.

OMG: optimized minimal 3D Gaussian splatting, enabling fast and efficient rendering with minimal storage.

Motion blur, usually a problem, is now a solution! This paper estimates camera motion from motion-blurred images, acting like an IMU.

FFaceNeRF: Enables few-shot face editing in NeRFs via geometry adapter & latent mixing, enhancing control & quality with limited training data.

Zero-1-to-A: Animatable avatars from a single image using video diffusion, robust to spatial & temporal inconsistencies!

Efficient image representation via adaptive token reduction.

VideoRFSplat: Direct text-to-3D Gaussian Splatting with flexible pose and multi-view joint modeling, bypassing SDS refinement!

FlashVDM enables fast 3D shape generation by accelerating both VAE decoding and diffusion sampling.

Uni-3DAR: Autoregressive framework unifies 3D generation/understanding, compressing spatial tokens for faster, versatile AI.

URA: Ultra-resolution adaptation made easy! Uses synthetic data & minor weight tuning for efficient, high-res text-to-image diffusion models.

TokenSet: Tokenizing images as unordered sets for dynamic capacity allocation and robust generation, breaking from fixed-position latent codes.

Sonata: Reliable 3D point cloud self-supervised learning through self-distillation, achieving SOTA with less data.

SWD: Scale-wise distillation of diffusion models achieves faster image generation by upscaling resolution during denoising, outperforming counterparts with similar computation.

SALT: Fine-tuning SAM for medical images using Singular Value Adaptation with Low-Rank Transformation for efficient, robust segmentation.

NuiScene: Enables efficient & unbounded outdoor scene generation by encoding scene chunks as uniform vector sets and outpainting.