Existing image classification datasets used in computer vision tend to have a uniform distribution of images across object categories. In contrast, the natural world is heavily imbalanced, as some species are more abundant and easier to photograph than others. To encourage further progress in challenging real world conditions we present the iNaturalist species classification and detection dataset, consisting of 859,000 images from over 5,000 different species of plants and animals. It features visually similar species, captured in a wide variety of situations, from all over the world. Images were collected with different camera types, have varying image quality, feature a large class imbalance, and have been verified by multiple citizen scientists. We discuss the collection of the dataset and present extensive baseline experiments using state-of-the-art computer vision classification and detection models. Results show that current non-ensemble based methods achieve only 67% top one classification accuracy, illustrating the difficulty of the dataset. Specifically, we observe poor results for classes with small numbers of training examples suggesting more attention is needed in low-shot learning.
DivShift: Exploring Domain-Specific Distribution Shifts in Large-Scale, Volunteer-Collected Biodiversity Datasets
Stanford University(斯坦福大学);University of Freiburg(弗莱堡大学);Carnegie Science(卡内基科学研究所);Federal University of Minas Gerais(米纳斯吉拉斯联邦大学);University of California, Berkeley(加州大学伯克利分校);Howard Hughes Medical Institute(霍华德·休斯医学研究所)
相机陷阱(或野生凸轮),使自动收集大量的图像数据。世界各地的生物学家都使用相机陷阱来监测动物物种的生物多样性和种群密度。计算机视觉社区一直在朝着自动化相机陷阱中的物种分类挑战迈进,但是当我们试图将这些模型的范围从我们收集训练数据的特定区域扩展到不同区域时,我们面临着一个有趣的问题:如何对新区域中的物种进行分类,而这些区域可能在以前的训练数据中没有看到?为了解决这个问题,我们准备了一个数据集和挑战,其中训练数据和测试数据来自不同的地区,即美国西南部和美国西北部。我们使用从美国西南部收集的Caltech Camera Traps数据集作为训练数据。我们添加了一个来自美国西北部的新数据集,该数据集由爱达荷州渔业和游戏部(IDFG)提供的数据整理而成,作为我们的测试数据集。测试数据与训练数据有一些类重叠,在两个数据集中都发现了一些物种,但在训练期间看到的两个物种在测试期间都没有看到,反之亦然。为了帮助填补训练物种的空白,我们允许竞争对手利用来自两个替代领域的迁移学习:iNaturalist的人工策划图像和Microsoft的TrapCam-AirSim模拟环境的合成图像。
Multispectral airborne laser scanning dataset for tree species classification: MS-ALS-SPECIES
Department of Remote Sensing and Photogrammetry(遥感与摄影测量系);Finnish Geospatial Research Institute(芬兰地理研究 institute);FGI;The National Land Survey of Finland(芬兰国家土地测绘局)
GlobalGeoTree: A Multi-Granular Vision-Language Dataset for Global Tree Species Classification
Technical University of Munich(慕尼黑技术大学);University of Reading(reading大学);University of Vienna(维也纳大学);University of Göttingen(哥廷根大学);University of Konstanz(康斯坦茨大学);Munich Center for Machine Learning(慕尼黑机器学习中心)
Expert Knowledge-Guided Decision Calibration for Accurate Fine-Grained Tree Species Classification
State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University(武汉大学测绘遥感信息工程国家重点实验室);Hubei Luojia Laboratory(湖北珞珈实验室)
Deep Learning for Automated Identification of Vietnamese Timber Species: A Tool for Ecological Monitoring and Conservation
Mathematical Modeling Laboratory, Graduate School of Bioresource and Bioenvironmental Sciences, Kyushu University(九州大学生物资源与生物环境科学研究生院数学建模实验室);International Cooperation Center, Thai Nguyen University(Thai Nguyen大学国际合作中心);Centrer of Crop Research for Adaptation to Climate Change, Thai Nguyen University of Agriculture and Forestry(Thai Nguyen农业与林业大学适应气候变化作物研究中心)
Multispectral airborne laser scanning for tree species classification: a benchmark of machine learning and deep learning algorithms
Department of Remote Sensing and Photogrammetry, Finnish Geospatial Research Institute FGI(遥感与摄影测量系,芬兰地理研究机构FGI);Department of Forest Sciences, University of Helsinki(森林科学系,赫尔辛基大学)
Benchmarking tree species classification from proximally-sensed laser scanning data: introducing the FOR-species20K dataset
Norwegian Institute for Bioeconomy Research (NIBIO)(挪威生物经济研究所);University of Cambridge(剑桥大学);Jan Evangelista Purkyně University in Ústí n. L.(拉贝河畔乌斯季扬·埃万杰利斯塔·普尔基涅大学);University of Freiburg(弗莱堡大学);Georg-August-Universität Göttingen(哥廷根大学);TU Wien(维也纳工业大学);Swiss Federal Institute for Forest, Snow and Landscape Research WSL(瑞士联邦森林、雪与景观研究所);University of British Columbia(不列颠哥伦比亚大学);Ghent University(根特大学);Heidelberg University(海德堡大学);University of Eastern Finland(东芬兰大学);Silva Tarouca Research Institute(席尔瓦·塔劳卡研究所);Forest Research Institute(森林研究所);Commonwealth Scientific and Industrial Research Organization (CSIRO)(联邦科学与工业研究组织);Charles Darwin University(查尔斯·达尔文大学);University College of London(伦敦大学学院);Natural Resources Institute Finland (Luke)(芬兰自然资源研究所);Council for Agricultural Research and Economics(农业研究与经济委员会);University of Lethbridge(莱斯布里奇大学);National Research Council – Institute of BioEconomy(国家研究委员会生物经济研究所);Free University of Bolzano(博尔扎诺自由大学)
Multi-species Seagrass Detection and Classification from Underwater Images
CSIRO Data61(英联邦科学与工业研究组织数据61部门);School of Electrical Engineering and Robotics, Queensland University of Technology (QUT)(昆士兰科技大学(QUT)电气工程与机器人学院);Queensland University of Technology (QUT)(昆士兰科技大学(QUT))
Lightweight Image Classification of Raptor Species for Edge Devices: Rare-Species Dataset Expansion via Video Frame Extraction, Knowledge Distillation, and TensorRT Deployment
Prospects for Mitigating Spectral Variability in Tropical Species Classification Using Self-Supervised Learning
AMAP;Univ Montpellier(蒙彼利埃大学);CIRAD(法国农业发展研究国际合作中心);CNRS(法国国家科学研究中心);INRAE(法国国家农业、食品与环境研究院);IRD(法国发展研究院);Technical University of Munich(慕尼黑工业大学);German Aerospace Center(德国航空航天中心);Univ. Grenoble Alpes(格勒诺布尔阿尔卑斯大学);Inria(法国国家信息与自动化研究所);Grenoble INP(格勒诺布尔理工学院);LJK(约瑟夫·傅里叶实验室)
From Articles to Canopies: Knowledge-Driven Pseudo-Labelling for Tree Species Classification using LLM Experts
Institute of Theoretical and Applied Informatics, Polish Academy of Sciences(波兰科学院理论与应用信息学研究所);University of Lodz(卢布林大学);University of Warsaw, Faculty of Geography and Regional Studies, Department of Geoinformatics, Cartography and Remote Sensing(华沙大学地理与区域研究学院,地理信息学、制图与遥感系)
NormalView: tree species classification from backpack and aerial lidar data using geometric projections
Finnish Geospatial Research Institute FGI(芬兰地理研究 institute FGI);The National Land Survey of Finland(芬兰国家土地调查局);Department of Remote Sensing and Photogrammetry(遥感与摄影测量系);Department of Built Environment, School of Engineering(环境系,工程学院)
Deep Learning Pipeline for Automated Visual Moth Monitoring: Insect Localization and Species Classification
Friedrich Schiller University Jena(耶拿弗里德里希·席勒大学);German Aerospace Center (DLR)(德国航空航天中心(DLR));Michael Stifel Center Jena for Data-Driven and Simulation Science(耶拿迈克尔·施蒂费尔数据驱动与模拟科学中心)
Bird Species Categorization Using Pose Normalized Deep Convolutional Nets
California Institute of Technology(加州理工学院);University of California, San Diego(加州大学圣迭戈分校);Cornell Tech(康奈尔科技校区)
我们提出了一个架构细粒度的视觉分类,方法专家人类的表现在鸟类的分类。我们的架构首先计算对象的姿态估计,这是用来计算本地图像的功能,反过来,用于分类。通过将深度卷积网络应用于由姿势定位和归一化的图像块来计算特征。我们进行了实证研究的姿态规范化计划,包括调查高阶几何扭曲功能。我们提出了一种新的基于图的聚类算法学习一个紧凑的姿势归一化空间。我们对最先进的深度卷积特征实现进行了详细的研究,并对细粒度分类的特征学习进行了微调。我们观察到,将较低级别的特征层与姿势归一化的提取例程和较高级别的特征层与未对齐的图像特征相结合的模型效果最好。我们的实验提高了鸟类物种识别的最新性能,与以前的方法相比,正确分类率有很大的提高(75% vs. 55-65%)。
A Novel Geographic Graph-Regulated Disentangling Mamba Model with Sparse Tokens for Enhanced Tree Species Classification from MODIS Time Series
Department of Geomatics Engineering, University of Calgary, Canada(卡尔加里大学测绘工程系);Department of Geography, Faculty of Arts, The University of Jordan(约旦大学艺术学院地理系)