Vocabulary-free fine-grained image recognition aims to distinguish visually similar categories within a meta-class without a fixed, human-defined label set. Exi...
Deep neural networks often exploit shortcuts. These are spurious cues which are associated with output labels in the training data but are unrelated to task sem...
Myth: Computer Vision is only effective for images and not for videos. Reality: Computer Vision can handle both images and videos, thanks to advancements in tem...
Handwritten text recognition and optical character recognition solutions show excellent results with processing data of modern era, but efficiency drops with La...
Introduction Person re-identification re-ID is the task of finding the same individual across different camera views. It has important applications in security...
Overview Teaching computers to recognize patterns without labeled data—known as unsupervised learning—has become more accessible thanks to simple tweaks to the...
Modern Latent Diffusion Models (LDMs) typically operate in low-level Variational Autoencoder (VAE) latent spaces that are primarily optimized for pixel-level re...
Monocular depth estimation remains challenging as recent foundation models, such as Depth Anything V2 (DA-V2), struggle with real-world images that are far from...
Recent progress in 3D reconstruction has made it easy to create realistic digital twins from everyday environments. However, current digital twins remain largel...
With the increase in deep learning, it becomes increasingly difficult to understand the model in which AI systems can identify objects. Thus, an adversary could...
Understanding and generating multi-person interactions is a fundamental challenge with broad implications for robotics and social computing. While humans natura...
We present RadarGen, a diffusion model for synthesizing realistic automotive radar point clouds from multi-view camera imagery. RadarGen adapts efficient image-...
Current approaches for designing self-explainable models (SEMs) require complicated training procedures and specific architectures which makes them impractical....
A key challenge in evaluating VLMs is testing models' ability to analyze visual content independently from their textual priors. Recent benchmarks such as BLINK...
Modern diffusion models (DMs) have achieved state-of-the-art image generation. However, the fundamental design choice of diffusing data all the way to white noi...
Plant diseases pose a significant threat to global food security, necessitating accurate and interpretable disease detection methods. This study introduces an i...
Text-to-image (T2I) diffusion models generate high-quality images but often fail to capture the spatial relations specified in text prompts. This limitation can...
We introduce PathBench-MIL, an open-source AutoML and benchmarking framework for multiple instance learning (MIL) in histopathology. The system automates end-to...
Depth-of-field control is essential in photography, but getting the perfect focus often takes several tries or special equipment. Single-image refocusing is sti...
We present WorldCanvas, a framework for promptable world events that enables rich, user-directed simulation by combining text, trajectories, and reference image...
Inspired by the success of generative pretraining in natural language, we ask whether the same principles can yield strong self-supervised visual learners. Inst...
Conventional evaluation methods for multimodal LLMs (MLLMs) lack interpretability and are often insufficient to fully disclose significant capability gaps acros...
Perceiving and reconstructing 3D scene geometry from visual inputs is crucial for autonomous driving. However, there still lacks a driving-targeted dense geomet...
While image editing has advanced rapidly, video editing remains less explored, facing challenges in consistency, control, and generalization. We study the desig...
Recent advances have shown that multimodal large language models (MLLMs) benefit from multimodal interleaved chain-of-thought (CoT) with vision tool interaction...
The rapid growth of stereoscopic displays, including VR headsets and 3D cinemas, has led to increasing demand for high-quality stereo video content. However, pr...
In this work, we present a panoramic metric depth foundation model that generalizes across diverse scene distances. We explore a data-in-the-loop paradigm from ...
Recent advances in multimodal models highlight the pivotal role of image tokenization in high-resolution image generation. By compressing images into compact la...
Prior works on 3D hand trajectory prediction are constrained by datasets that decouple motion from semantic supervision and by models that weakly link reasoning...
Reward models (RMs) are essential for training large language models (LLMs), but remain underexplored for omni models that handle interleaved image and text seq...
Video Large Language Models (VLLMs) unlock world-knowledge-aware video understanding through pretraining on internet-scale data and have already shown promise o...
The application of Machine Learning (ML) to the diagnosis of rare diseases, such as collagen VI-related dystrophies (COL6-RD), is fundamentally limited by the s...
Existing video generation models struggle to maintain long-term spatial and temporal consistency due to the dense, high-dimensional nature of video signals. To ...
At the most basic level, pixels are the source of the visual information through which we perceive the world. Pixels contain information at all levels, ranging ...
In recent multimodal research, the diffusion paradigm has emerged as a promising alternative to the autoregressive paradigm (AR), owing to its unique decoding a...
We present Gaussian Pixel Codec Avatars (GPiCA), photorealistic head avatars that can be generated from multi-view images and efficiently rendered on mobile dev...
Foundation models are vital tools in various Computer Vision applications. They take as input a single RGB image and output a deep feature representation that i...
Active Speaker Detection (ASD) aims to identify who is currently speaking in each frame of a video. Most state-of-the-art approaches rely on late fusion to comb...
Autoregressive video diffusion models hold promise for world simulation but are vulnerable to exposure bias arising from the train-test mismatch. While recent w...
Evaluations of image compression performance which include human preferences have generally found that naive distortion functions such as MSE are insufficiently...
The misuse of AI-driven video generation technologies has raised serious social concerns, highlighting the urgent need for reliable AI-generated video detectors...
Prevailing Vision-Language-Action Models (VLAs) for robotic manipulation are built upon vision-language backbones pretrained on large-scale, but disconnected st...
This paper proposes a training data augmentation pipeline that combines synthetic image data with neural style transfer in order to address the vulnerability of...
The computational and memory overheads associated with expanding the context window of LLMs severely limit their scalability. A noteworthy solution is vision-te...
Working memory enables the brain to integrate transient information for rapid decision-making. Artificial networks typically replicate this via recurrent or par...
Edit your config.toml while the app is running and watch the pipeline update instantly. No recompiling. No stopping the camera. Pure iteration bliss. Why This M...
Introduction Data annotation is a foundational process in artificial intelligence that enables machines to learn from real‑world data. It involves adding meani...
An AI background remover may feel like magic at first glance. You upload an image, click a button, and the background disappears. Behind that simple interaction...
