Machine Learning and Optimization -- Sublabel-Accurate Multilabeling Meets Product Label Spaces -- InfoSeg: Unsupervised Semantic Image Segmentation with Mutual Information Maximization -- Sampling-free Variational Inference for Neural Networks with Multiplicative Activation Noise -- Conditional Adversarial Debiasing: Towards Learning Unbiased Classifiers from Biased Data -- Revisiting Consistency Regularization for Semi-Supervised Learning -- Learning Robust Models Using the Principle of Independent Causal Mechanisms -- Reintroducing Straight-Through Estimators as Principled Methods for Stochastic Binary Networks -- Bias-Variance Tradeoffs in Single-Sample Binary Gradient Estimators -- End-to-end Learning of Fisher Vector Encodings for Part Features in Fine-grained Recognition -- Investigating the Consistency of Uncertainty Sampling in Deep Active Learning -- ScaleNet: An Unsupervised Representation Learning Method for Limited Information -- Actions, Events, and Segmentation -- A New Split for Evaluating True Zero-Shot Action Recognition -- Video Instance Segmentation with Recurrent Graph Neural Networks -- Distractor-Aware Video Object Segmentation -- (SP)^2Net for Generalized Zero-Label Semantic Segmentation -- Contrastive Representation Learning for Hand Shape Estimation -- Fusion-GCN: Multimodal Action Recognition using Graph Convolutional Networks -- FIFA: Fast Inference Approximation for Action Segmentation -- Hybrid SNN-ANN: Energy-Efficient Classification and Object Detection for Event-Based Vision -- A Comparative Study of PnP and Learning Approaches to Super-Resolution in a Real-World Setting -- Merging-ISP: Multi-Exposure High Dynamic Range Image Signal Processing -- Spatiotemporal Outdoor Lighting Aggregation on Image Sequences -- Generative Models and Multimodal Data -- AttrLostGAN: Attribute Controlled Image Synthesis from Reconfigurable Layout and Style -- Learning Conditional Invariance through Cycle Consistency -- CAGAN: Text-To-Image Generation with Combined Attention Generative Adversarial Networks -- TxT: Crossmodal End-to-End Learning with Transformers -- Diverse Image Captioning with Grounded Style -- Labeling and Self-Supervised Learning -- Leveraging Group Annotations in Object Detection Using Graph-Based Pseudo-Labeling -- Quantifying Uncertainty of Image Labelings Using Assignment Flows -- Implicit and Explicit Attention for Zero-Shot Learning -- Self-Supervised Learning for Object Detection in Autonomous Driving -- Assignment Flows and Nonlocal PDEs on Graphs -- Applications -- Viewpoint-Tolerant Semantic Segmentation for Aerial Logistics -- T6D-Direct: Transformers for Multi-Object 6D Pose Direct Regression -- TetraPackNet: Four-Corner-Based Object Detection in Logistics Use-Cases -- Detecting Slag Formations with Deep Convolutional Neural Networks -- Virtual Temporal Samples for Recurrent Neural Networks: applied to semantic segmentation in agriculture -- Weakly Supervised Segmentation Pre-training for Plant Cover Prediction -- How Reliable Are Out-of-Distribution Generalization Methods for Medical Image Segmentation? -- 3D Modeling and Reconstruction -- Clustering Persistent Scatterer Points Based on a Hybrid Distance Metric -- CATEGORISE: An Automated Framework for Utilizing the Workforce of the Crowd for Semantic Segmentation of 3D Point Clouds -- Zero-Shot remote sensing image super resolution based on image continuity and self-tessellations -- A Comparative Survey of Geometric Light Source Calibration Methods -- Quantifying point cloud realism through adversarially learned latent representations -- Full-Glow: Fully conditional Glow for more realistic image generation -- Multidirectional Conjugate Gradients for Scalable Bundle Adjustment. .

This book constitutes the refereed proceedings of the 43rd DAGM German Conference on Pattern Recognition, DAGM GCPR 2021, which was held during September 28 - October 1, 2021. The conference was planned to take place in Bonn, Germany, but changed to a virtual event due to the COVID-19 pandemic. The 46 papers presented in this volume were carefully reviewed and selected from 116 submissions. They were organized in topical sections as follows: machine learning and optimization; actions, events, and segmentation; generative models and multimodal data; labeling and self-supervised learning; applications; and 3D modelling and reconstruction.