Browse research papers in Computer Vision and Pattern Recognition - Computer Science Discover 25 academic papers with AI-powered analysis

A SCENE IS WORTH A THOUSAND FEATURES: FEED-FORWARD CAMERA LOCALIZATION FROM A COLLECTION OF IMAGE FEATURES

This paper introduces FastForward, a novel computer vision method for quickly and accurately determining a camera's exact location and orientation in a 3D scene. By representing scenes as a sparse collection of image features and using a single feed-forward neural network pass, FastForward significantly reduces the time and resources required for mapping a scene while achieving state-of-the-art or comparable accuracy to existing methods across diverse indoor and outdoor environments. The approach also demonstrates robust generalization to unseen domains and varying scale ranges thanks to a scene and scale normalization technique.

★

☆

paper_2167.pdf

Oct 02, 02:36 PM

CMT: MID-TRAINING FOR EFFICIENT LEARNING OFCONSISTENCY, MEAN FLOW, AND FLOW MAP MODELS

The paper introduces Consistency Mid-Training (CMT), a novel intermediate training stage designed to significantly improve the efficiency, stability, and performance of flow map models for vision generation. CMT acts as a bridge between pre-training (diffusion models) and post-training (flow map models), providing a trajectory-consistent initialization that reduces total training cost (data and GPU time) by up to 98% compared to baselines, while achieving state-of-the-art FID scores on various image generation benchmarks. The theoretical analysis confirms that CMT provides a strong starting point for flow map post-training, minimizing gradient bias and accelerating convergence.

Understanding the Effective Receptive Field in Deep Convolutional Neural Networks

This paper introduces the concept of an "effective receptive field" (ERF) in deep convolutional neural networks (CNNs), showing it's smaller than the theoretical receptive field and follows a Gaussian distribution. The authors analyze how ERF size is affected by factors like network depth, non-linear activations, and skip connections, and they suggest ways to increase ERF size, such as modified initialization and architectural changes.

Tutorial on Diffusion Models for Imaging and Vision

This tutorial provides a comprehensive overview of diffusion models, tracing their development from variational autoencoders (VAEs) to denoising diffusion probabilistic models (DDPMs) and score-matching Langevin dynamics (SMLD). It also explores the connection between diffusion models and stochastic differential equations (SDEs), providing insights into their underlying principles and behavior.

Sensory Robustness through Top-Down Feedback and Neural Stochasticity in Recurrent Vision Models

This study explored how top-down feedback and simulated neural noise (dropout) affect the performance of convolutional recurrent neural networks (ConvRNNs) on image classification. They found that only when both top-down feedback and dropout were combined, the ConvRNNs became more robust to noisy or manipulated images, outperforming models with either feature alone.

Registration beyond Points: General Affine Subspace Alignment via Geodesic Distance on Grassmann Manifold

This paper proposes a new method for registering lines and planes in 3D by minimizing the geodesic distance on the Grassmann manifold, which offers a more theoretically sound and robust approach compared to existing methods that rely on Euclidean distances or point approximations. Experimental results on object registration, RGB-D odometry, and camera pose estimation demonstrate improved accuracy and convergence, especially in the presence of outliers.

Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs

This paper introduces Cambrian-1, a family of open-source multimodal large language models (MLLMs) focused on improving visual understanding. Cambrian-1 achieves state-of-the-art performance on several benchmarks, matching or exceeding some proprietary models. The authors also develop a new vision-centric benchmark and propose a more efficient connector design for vision and language integration.

Meta CLIP 2: A Worldwide Scaling Recipe

This paper introduces Meta CLIP 2, a new model trained on a massive dataset of image-text pairs from various languages, resulting in improved performance on both English and multilingual tasks. The key innovation is a scaling recipe involving metadata, curation, and training capacity adjustments. The model achieves state-of-the-art results on several multilingual benchmarks, including XM3600, Babel-ImageNet, and CVQA.

Streaming 4D Visual Geometry Transformer

This paper introduces StreamVGGT, a causal transformer model that reconstructs 4D spatial-temporal geometry from video in real-time. By caching historical tokens and using causal attention, it processes video frames incrementally, offering faster inference than traditional methods while maintaining competitive accuracy thanks to knowledge distillation from a more computationally expensive teacher model.

Segment Anything

This paper introduces the Segment Anything Model (SAM), a promptable segmentation model capable of generating masks from various input prompts like points, boxes, and text. SAM is trained on SA-1B, a massive dataset containing over 1 billion masks, enabling impressive zero-shot transfer capabilities to a diverse range of segmentation tasks. The authors demonstrate SAM's effectiveness through experiments on edge detection, object proposal generation, instance segmentation, and text-to-mask prediction.

Action Transformer: A Self-Attention Model for Short-Time Pose-Based Human Action Recognition

The Action Transformer (AcT), a purely self-attentional model, excels at recognizing short-time human actions from 2D pose data. Outperforming previous methods on a new dataset, MPOSE2021, AcT also shows promise for low-latency, real-time applications due to its efficient design.

Fine-Grained Image Analysis with Deep Learning: A Survey

The survey examines deep learning advancements in fine-grained image analysis (FGIA), arguing for a broader definition encompassing recognition and retrieval. It presents a taxonomy of techniques, evaluates performance on benchmarks, and outlines future research directions, including precise definition, new datasets, 3D application, robust representations, and interpretability.

★

☆

Fine_Grained_Image.pdf

Jul 14, 05:20 PM

Object Detection in Aerial Images: A Large-Scale Benchmark and Challenges

This paper introduces DOTA, a massive dataset for object detection in aerial images, featuring 1.8 million object instances across 18 categories with oriented bounding box annotations. Using this dataset, they benchmark 10 state-of-the-art object detection algorithms across 70+ configurations, providing a valuable resource for researchers in the field and demonstrating the unique challenges of aerial object detection.

Face mask detection using YOLOv3 and faster R-CNN models: COVID-19 environment

This paper compares YOLOv3 and Faster R-CNN for real-time face mask detection. Faster R-CNN had slightly better accuracy, while YOLOv3 significantly outperformed in speed, suggesting YOLOv3 is preferable for real-time applications like surveillance.

★

☆

s11042-021-10711-8.pdf

Jul 14, 05:20 PM

Computer Vision and Pattern Recognition 2020

This editorial summarizes the top four papers from the 42nd German Conference on Pattern Recognition (DAGM GCPR 2020), covering camera calibration, OCT segmentation, video representation learning, and the interpretability of deep neural networks. These papers offer insights into the current state and future directions of computer vision research.

★

☆

s11263-021-01522-3.pdf

Jul 14, 05:20 PM

Masked Autoencoders Are Scalable Vision Learners

This paper introduces Masked Autoencoders (MAE), a self-supervised learning approach for computer vision. By masking large portions of an image and training a model to reconstruct the missing parts, MAE learns highly effective visual representations that achieve state-of-the-art results on ImageNet and improve transfer learning performance on various downstream tasks.

High-Resolution Image Synthesis with Latent Diffusion Models

This paper introduces Latent Diffusion Models (LDMs), a new approach to image synthesis that reduces the computational demands of traditional diffusion models while maintaining high-quality results. By operating in the latent space of a pre-trained autoencoder, LDMs achieve faster training and sampling while also enabling flexible conditioning on various inputs like text or bounding boxes.

Swin Transformer: Hierarchical Vision Transformer using Shifted Windows

This paper proposes a novel vision Transformer architecture called Swin Transformer which utilizes a shifted window approach for computing self-attention, resulting in linear computational complexity. Experiments on ImageNet, COCO, and ADE20K datasets demonstrate state-of-the-art performance across image classification, object detection, and semantic segmentation tasks.

YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors

YOLOv7 introduces architectural improvements and training optimizations, called "trainable bag-of-freebies," to boost the speed and accuracy of real-time object detection. The different versions of YOLOv7 achieve state-of-the-art results on the COCO dataset across a range of FPS.

Semantic interpretation of architectural and archaeological geometries: Point cloud segmentation for HBIM parameterisation

This paper applies Brodu and Lague's morphological segmentation algorithm (CANUPO) to classify architectural components from terrestrial laser scanning data of a historical palace façade. The algorithm performed well on large complex shapes but struggled with smaller details, suggesting its applicability for general façade analysis but limitations for fine-grained modeling.

★

☆

Manuscrip_Prepint_merged.pdf

Jul 14, 05:09 PM

Computer Vision and Pattern Recognition

Paperzilla Papers in Computer Vision and Pattern Recognition

A SCENE IS WORTH A THOUSAND FEATURES: FEED-FORWARD CAMERA LOCALIZATION FROM A COLLECTION OF IMAGE FEATURES

CMT: MID-TRAINING FOR EFFICIENT LEARNING OFCONSISTENCY, MEAN FLOW, AND FLOW MAP MODELS

Understanding the Effective Receptive Field in Deep Convolutional Neural Networks

Tutorial on Diffusion Models for Imaging and Vision

Sensory Robustness through Top-Down Feedback and Neural Stochasticity in Recurrent Vision Models

B-VLLM: A Vision Large Language Model with Balanced Spatio-Temporal Tokens

MobileCLIP2: Improving Multi-Modal Reinforced Training

DensePose From WiFi

Self-Rewarding Vision-Language Model via Reasoning Decomposition

HAMSt3R: Human-Aware Multi-view Stereo 3D Reconstruction

Registration beyond Points: General Affine Subspace Alignment via Geodesic Distance on Grassmann Manifold

Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs

Meta CLIP 2: A Worldwide Scaling Recipe

Streaming 4D Visual Geometry Transformer

Segment Anything

Action Transformer: A Self-Attention Model for Short-Time Pose-Based Human Action Recognition

Fine-Grained Image Analysis with Deep Learning: A Survey

Object Detection in Aerial Images: A Large-Scale Benchmark and Challenges

Face mask detection using YOLOv3 and faster R-CNN models: COVID-19 environment

Computer Vision and Pattern Recognition 2020

Masked Autoencoders Are Scalable Vision Learners

High-Resolution Image Synthesis with Latent Diffusion Models

Swin Transformer: Hierarchical Vision Transformer using Shifted Windows

YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors

Semantic interpretation of architectural and archaeological geometries: Point cloud segmentation for HBIM parameterisation