Dr. Antonio M. López is the head of the Advanced Driver Assistance Systems (ADAS) Group of the Computer Vision Center (CVC), and Associate Professor of the Computer Science Department, both from the Universitat Autònoma de Barcelona (UAB).  Antonio received a BSc degree in Computer Science from the Universitat Politècnica de Catalunya (UPC) and a PhD degree in Computer Vision from the Universitat Autònoma de Barcelona (UAB). In 1996, he participated in the foundation of the CVC at the UAB, where he has held different institutional responsibilities. Antonio is also the responsible of the Software Engineering specialty at the UAB. Moreover, he has been the principal investigator of numerous public and industrial research projects, and is a co-author of more than 100 journal and conference papers, all in the field of computer vision. Antonio's main research interests are vision-based driver assistance and autonomous driving.Atsushi Imiya is Professor at IMIT, Chiba University. He has served as a PC member of DGCI, IWCIA, and SSVM conferences for many years. He is an editorial member of “Pattern Recognition (Journal)” and a co-editor of “Digital and Image Geometry” held at Schloss Dagstuhl in 2000, MLDM2007 (Machine Learning and Data Mining in Pattern Recognition), of which proceedings were published from Springer-Verlag. He is a general co-chair of S+SSPR (Statistical, and Synthetic and Structural Pattern Recognition) 2012. He is participating in a government-funded project titled: “Computational anatomy for computer-aided diagnosis and therapy: Frontiers of medical image sciences” as an applied mathematician. He also serves as a review committee of the research projects internationally.Dr. Tomas Pajdla is an Assistant Professor and Distinguished Senior Researcher at the Czech Technical University in Prague. He works in geometry and algebra of computer vision and robotics with the emphasis on geometry a calibration of camera systems, 3D reconstruction and industrial vision. Dr. Pajdla published more than 75 works in journals and proceedings and received awards for his work; OAGM 1998, 2012, BMVC 2002, ICCV 2005 and ACCV 2014. He has served as a program co-chair of ECCV 2004 and ECCV 2014, and regularly as area chair of ICCV, CVPR, ECCV, ACCV, ICRA and BMVC. He is a member of the ECCV Board, and served on the boards of IEEE PAMI, Computer Vision and Image Understanding and IPSJ Transactions on Computer Vision and Applications journals. Dr. Pajdla has connections to the planetary research community through EU projects with NASA, ESA and EADS Astrium and to automotive industry via Daimler AG.Jose M. Alvarez is currently a researcher at NICTA and a research fellow at the Australian National University, Canberra, Australia. Previously, he was a postdoctoral researcher at the Computational and Biological Learning Group at New York University with Professor Yann LeCun. During his Ph.D. he was a visiting researcher at the University of Amsterdam and Volkswagen AG research. His main research interests include deep learning and data driven methods for dynamic scene understanding.

• List of Contributors ixPreface xiAbbreviations and Acronyms xiii1 Computer Vision in Vehicles 1Reinhard Klette1.1 Adaptive Computer Vision for Vehicles 11.1.1 Applications 11.1.2 Traffic Safety and Comfort 21.1.3 Strengths of (Computer) Vision 21.1.4 Generic and Specific Tasks 31.1.5 Multi-module Solutions 41.1.6 Accuracy, Precision, and Robustness 51.1.7 Comparative Performance Evaluation 51.1.8 There Are Many Winners 61.2 Notation and Basic Definitions 61.2.1 Images and Videos 61.2.2 Cameras 81.2.3 Optimization 101.3 Visual Tasks 121.3.1 Distance 121.3.2 Motion 161.3.3 Object Detection and Tracking 181.3.4 Semantic Segmentation 211.4 Concluding Remarks 23Acknowledgments 232 Autonomous Driving 24Uwe Franke2.1 Introduction 242.1.1 The Dream 242.1.2 Applications 252.1.3 Level of Automation 262.1.4 Important Research Projects 272.1.5 Outdoor Vision Challenges 302.2 Autonomous Driving in Cities 312.2.1 Localization 332.2.2 Stereo Vision-Based Perception in 3D 362.2.3 Object Recognition 432.3 Challenges 492.3.1 Increasing Robustness 492.3.2 Scene Labeling 502.3.3 Intention Recognition 522.4 Summary 52Acknowledgments 543 Computer Vision for MAVs 55Friedrich Fraundorfer3.1 Introduction 553.2 System and Sensors 573.3 Ego-Motion Estimation 583.3.1 State Estimation Using Inertial and Vision Measurements 583.3.2 MAV Pose from Monocular Vision 623.3.3 MAV Pose from Stereo Vision 633.3.4 MAV Pose from Optical Flow Measurements 653.4 3D Mapping 673.5 Autonomous Navigation 713.6 Scene Interpretation 723.7 Concluding Remarks 734 Exploring the Seafloor with Underwater Robots 75Rafael Garcia, Nuno Gracias, Tudor Nicosevici, Ricard Prados, Natalia Hurtos, Ricard Campos, Javier Escartin, Armagan Elibol, Ramon Hegedus and Laszlo Neumann4.1 Introduction 754.2 Challenges of Underwater Imaging 774.3 Online Computer Vision Techniques 794.3.1 Dehazing 794.3.2 Visual Odometry 844.3.3 SLAM 874.3.4 Laser Scanning 914.4 Acoustic Imaging Techniques 924.4.1 Image Formation 924.4.2 Online Techniques for Acoustic Processing 954.5 Concluding Remarks 98Acknowledgments 995 Vision-Based Advanced Driver Assistance Systems 100David Gerónimo, David Vázquez and Arturo de la Escalera5.1 Introduction 1005.2 Forward Assistance 1015.2.1 Adaptive Cruise Control (ACC) and Forward Collision Avoidance (FCA) 1015.2.2 Traffic Sign Recognition (TSR) 1035.2.3 Traffic Jam Assist (TJA) 1055.2.4 Vulnerable Road User Protection 1065.2.5 Intelligent Headlamp Control 1095.2.6 Enhanced Night Vision (Dynamic Light Spot) 1105.2.7 Intelligent Active Suspension 1115.3 Lateral Assistance 1125.3.1 Lane Departure Warning (LDW) and Lane Keeping System (LKS) 1125.3.2 Lane Change Assistance (LCA) 1155.3.3 Parking Assistance 1165.4 Inside Assistance 1175.4.1 Driver Monitoring and Drowsiness Detection 1175.5 Conclusions and Future Challenges 1195.5.1 Robustness 1195.5.2 Cost 121Acknowledgments 1216 Application Challenges from a Bird’s-Eye View 122Davide Scaramuzza6.1 Introduction to Micro Aerial Vehicles (MAVs) 1226.1.1 Micro Aerial Vehicles (MAVs) 1226.1.2 Rotorcraft MAVs 1236.2 GPS-Denied Navigation 1246.2.1 Autonomous Navigation with Range Sensors 1246.2.2 Autonomous Navigation with Vision Sensors 1256.2.3 SFLY: Swarm of Micro Flying Robots 1266.2.4 SVO, a Visual-Odometry Algorithm for MAVs 1266.3 Applications and Challenges 1276.3.1 Applications 1276.3.2 Safety and Robustness 1286.4 Conclusions 1327 Application Challenges of Underwater Vision 133Nuno Gracias, Rafael Garcia, Ricard Campos, Natalia Hurtos, Ricard Prados, ASM Shihavuddin, Tudor Nicosevici, Armagan Elibol, Laszlo Neumann and Javier Escartin7.1 Introduction 1337.2 Offline Computer Vision Techniques for Underwater Mapping and Inspection 1347.2.1 2D Mosaicing 1347.2.2 2.5D Mapping 1447.2.3 3D Mapping 1467.2.4 Machine Learning for Seafloor Classification 1547.3 Acoustic Mapping Techniques 1577.4 Concluding Remarks 1598 Closing Notes 161Antonio M. LópezReferences 164Index 195

Dr. Antonio M. López is the head of the Advanced Driver Assistance Systems (ADAS) Group of the Computer Vision Center (CVC), and Associate Professor of the Computer Science Department, both from the Universitat Autònoma de Barcelona (UAB).  Antonio received a BSc degree in Computer Science from the Universitat Politècnica de Catalunya (UPC) and a PhD degree in Computer Vision from the Universitat Autònoma de Barcelona (UAB). In 1996, he participated in the foundation of the CVC at the UAB, where he has held different institutional responsibilities. Antonio is also the responsible of the Software Engineering specialty at the UAB. Moreover, he has been the principal investigator of numerous public and industrial research projects, and is a co-author of more than 100 journal and conference papers, all in the field of computer vision. Antonio's main research interests are vision-based driver assistance and autonomous driving.Atsushi Imiya is Professor at IMIT, Chiba University. He has served as a PC member of DGCI, IWCIA, and SSVM conferences for many years. He is an editorial member of “Pattern Recognition (Journal)” and a co-editor of “Digital and Image Geometry” held at Schloss Dagstuhl in 2000, MLDM2007 (Machine Learning and Data Mining in Pattern Recognition), of which proceedings were published from Springer-Verlag. He is a general co-chair of S+SSPR (Statistical, and Synthetic and Structural Pattern Recognition) 2012. He is participating in a government-funded project titled: “Computational anatomy for computer-aided diagnosis and therapy: Frontiers of medical image sciences” as an applied mathematician. He also serves as a review committee of the research projects internationally.Dr. Tomas Pajdla is an Assistant Professor and Distinguished Senior Researcher at the Czech Technical University in Prague. He works in geometry and algebra of computer vision and robotics with the emphasis on geometry a calibration of camera systems, 3D reconstruction and industrial vision. Dr. Pajdla published more than 75 works in journals and proceedings and received awards for his work; OAGM 1998, 2012, BMVC 2002, ICCV 2005 and ACCV 2014. He has served as a program co-chair of ECCV 2004 and ECCV 2014, and regularly as area chair of ICCV, CVPR, ECCV, ACCV, ICRA and BMVC. He is a member of the ECCV Board, and served on the boards of IEEE PAMI, Computer Vision and Image Understanding and IPSJ Transactions on Computer Vision and Applications journals. Dr. Pajdla has connections to the planetary research community through EU projects with NASA, ESA and EADS Astrium and to automotive industry via Daimler AG.Jose M. Alvarez is currently a researcher at NICTA and a research fellow at the Australian National University, Canberra, Australia. Previously, he was a postdoctoral researcher at the Computational and Biological Learning Group at New York University with Professor Yann LeCun. During his Ph.D. he was a visiting researcher at the University of Amsterdam and Volkswagen AG research. His main research interests include deep learning and data driven methods for dynamic scene understanding.

• List of Contributors ixPreface xiAbbreviations and Acronyms xiii1 Computer Vision in Vehicles 1Reinhard Klette1.1 Adaptive Computer Vision for Vehicles 11.1.1 Applications 11.1.2 Traffic Safety and Comfort 21.1.3 Strengths of (Computer) Vision 21.1.4 Generic and Specific Tasks 31.1.5 Multi-module Solutions 41.1.6 Accuracy, Precision, and Robustness 51.1.7 Comparative Performance Evaluation 51.1.8 There Are Many Winners 61.2 Notation and Basic Definitions 61.2.1 Images and Videos 61.2.2 Cameras 81.2.3 Optimization 101.3 Visual Tasks 121.3.1 Distance 121.3.2 Motion 161.3.3 Object Detection and Tracking 181.3.4 Semantic Segmentation 211.4 Concluding Remarks 23Acknowledgments 232 Autonomous Driving 24Uwe Franke2.1 Introduction 242.1.1 The Dream 242.1.2 Applications 252.1.3 Level of Automation 262.1.4 Important Research Projects 272.1.5 Outdoor Vision Challenges 302.2 Autonomous Driving in Cities 312.2.1 Localization 332.2.2 Stereo Vision-Based Perception in 3D 362.2.3 Object Recognition 432.3 Challenges 492.3.1 Increasing Robustness 492.3.2 Scene Labeling 502.3.3 Intention Recognition 522.4 Summary 52Acknowledgments 543 Computer Vision for MAVs 55Friedrich Fraundorfer3.1 Introduction 553.2 System and Sensors 573.3 Ego-Motion Estimation 583.3.1 State Estimation Using Inertial and Vision Measurements 583.3.2 MAV Pose from Monocular Vision 623.3.3 MAV Pose from Stereo Vision 633.3.4 MAV Pose from Optical Flow Measurements 653.4 3D Mapping 673.5 Autonomous Navigation 713.6 Scene Interpretation 723.7 Concluding Remarks 734 Exploring the Seafloor with Underwater Robots 75Rafael Garcia, Nuno Gracias, Tudor Nicosevici, Ricard Prados, Natalia Hurtos, Ricard Campos, Javier Escartin, Armagan Elibol, Ramon Hegedus and Laszlo Neumann4.1 Introduction 754.2 Challenges of Underwater Imaging 774.3 Online Computer Vision Techniques 794.3.1 Dehazing 794.3.2 Visual Odometry 844.3.3 SLAM 874.3.4 Laser Scanning 914.4 Acoustic Imaging Techniques 924.4.1 Image Formation 924.4.2 Online Techniques for Acoustic Processing 954.5 Concluding Remarks 98Acknowledgments 995 Vision-Based Advanced Driver Assistance Systems 100David Gerónimo, David Vázquez and Arturo de la Escalera5.1 Introduction 1005.2 Forward Assistance 1015.2.1 Adaptive Cruise Control (ACC) and Forward Collision Avoidance (FCA) 1015.2.2 Traffic Sign Recognition (TSR) 1035.2.3 Traffic Jam Assist (TJA) 1055.2.4 Vulnerable Road User Protection 1065.2.5 Intelligent Headlamp Control 1095.2.6 Enhanced Night Vision (Dynamic Light Spot) 1105.2.7 Intelligent Active Suspension 1115.3 Lateral Assistance 1125.3.1 Lane Departure Warning (LDW) and Lane Keeping System (LKS) 1125.3.2 Lane Change Assistance (LCA) 1155.3.3 Parking Assistance 1165.4 Inside Assistance 1175.4.1 Driver Monitoring and Drowsiness Detection 1175.5 Conclusions and Future Challenges 1195.5.1 Robustness 1195.5.2 Cost 121Acknowledgments 1216 Application Challenges from a Bird’s-Eye View 122Davide Scaramuzza6.1 Introduction to Micro Aerial Vehicles (MAVs) 1226.1.1 Micro Aerial Vehicles (MAVs) 1226.1.2 Rotorcraft MAVs 1236.2 GPS-Denied Navigation 1246.2.1 Autonomous Navigation with Range Sensors 1246.2.2 Autonomous Navigation with Vision Sensors 1256.2.3 SFLY: Swarm of Micro Flying Robots 1266.2.4 SVO, a Visual-Odometry Algorithm for MAVs 1266.3 Applications and Challenges 1276.3.1 Applications 1276.3.2 Safety and Robustness 1286.4 Conclusions 1327 Application Challenges of Underwater Vision 133Nuno Gracias, Rafael Garcia, Ricard Campos, Natalia Hurtos, Ricard Prados, ASM Shihavuddin, Tudor Nicosevici, Armagan Elibol, Laszlo Neumann and Javier Escartin7.1 Introduction 1337.2 Offline Computer Vision Techniques for Underwater Mapping and Inspection 1347.2.1 2D Mosaicing 1347.2.2 2.5D Mapping 1447.2.3 3D Mapping 1467.2.4 Machine Learning for Seafloor Classification 1547.3 Acoustic Mapping Techniques 1577.4 Concluding Remarks 1598 Closing Notes 161Antonio M. LópezReferences 164Index 195

Dr. Antonio M. López is the head of the Advanced Driver Assistance Systems (ADAS) Group of the Computer Vision Center (CVC), and Associate Professor of the Computer Science Department, both from the Universitat Autònoma de Barcelona (UAB).  Antonio received a BSc degree in Computer Science from the Universitat Politècnica de Catalunya (UPC) and a PhD degree in Computer Vision from the Universitat Autònoma de Barcelona (UAB). In 1996, he participated in the foundation of the CVC at the UAB, where he has held different institutional responsibilities. Antonio is also the responsible of the Software Engineering specialty at the UAB. Moreover, he has been the principal investigator of numerous public and industrial research projects, and is a co-author of more than 100 journal and conference papers, all in the field of computer vision. Antonio's main research interests are vision-based driver assistance and autonomous driving.Atsushi Imiya is Professor at IMIT, Chiba University. He has served as a PC member of DGCI, IWCIA, and SSVM conferences for many years. He is an editorial member of “Pattern Recognition (Journal)” and a co-editor of “Digital and Image Geometry” held at Schloss Dagstuhl in 2000, MLDM2007 (Machine Learning and Data Mining in Pattern Recognition), of which proceedings were published from Springer-Verlag. He is a general co-chair of S+SSPR (Statistical, and Synthetic and Structural Pattern Recognition) 2012. He is participating in a government-funded project titled: “Computational anatomy for computer-aided diagnosis and therapy: Frontiers of medical image sciences” as an applied mathematician. He also serves as a review committee of the research projects internationally.Dr. Tomas Pajdla is an Assistant Professor and Distinguished Senior Researcher at the Czech Technical University in Prague. He works in geometry and algebra of computer vision and robotics with the emphasis on geometry a calibration of camera systems, 3D reconstruction and industrial vision. Dr. Pajdla published more than 75 works in journals and proceedings and received awards for his work; OAGM 1998, 2012, BMVC 2002, ICCV 2005 and ACCV 2014. He has served as a program co-chair of ECCV 2004 and ECCV 2014, and regularly as area chair of ICCV, CVPR, ECCV, ACCV, ICRA and BMVC. He is a member of the ECCV Board, and served on the boards of IEEE PAMI, Computer Vision and Image Understanding and IPSJ Transactions on Computer Vision and Applications journals. Dr. Pajdla has connections to the planetary research community through EU projects with NASA, ESA and EADS Astrium and to automotive industry via Daimler AG.Jose M. Alvarez is currently a researcher at NICTA and a research fellow at the Australian National University, Canberra, Australia. Previously, he was a postdoctoral researcher at the Computational and Biological Learning Group at New York University with Professor Yann LeCun. During his Ph.D. he was a visiting researcher at the University of Amsterdam and Volkswagen AG research. His main research interests include deep learning and data driven methods for dynamic scene understanding.

• List of Contributors ixPreface xiAbbreviations and Acronyms xiii1 Computer Vision in Vehicles 1Reinhard Klette1.1 Adaptive Computer Vision for Vehicles 11.1.1 Applications 11.1.2 Traffic Safety and Comfort 21.1.3 Strengths of (Computer) Vision 21.1.4 Generic and Specific Tasks 31.1.5 Multi-module Solutions 41.1.6 Accuracy, Precision, and Robustness 51.1.7 Comparative Performance Evaluation 51.1.8 There Are Many Winners 61.2 Notation and Basic Definitions 61.2.1 Images and Videos 61.2.2 Cameras 81.2.3 Optimization 101.3 Visual Tasks 121.3.1 Distance 121.3.2 Motion 161.3.3 Object Detection and Tracking 181.3.4 Semantic Segmentation 211.4 Concluding Remarks 23Acknowledgments 232 Autonomous Driving 24Uwe Franke2.1 Introduction 242.1.1 The Dream 242.1.2 Applications 252.1.3 Level of Automation 262.1.4 Important Research Projects 272.1.5 Outdoor Vision Challenges 302.2 Autonomous Driving in Cities 312.2.1 Localization 332.2.2 Stereo Vision-Based Perception in 3D 362.2.3 Object Recognition 432.3 Challenges 492.3.1 Increasing Robustness 492.3.2 Scene Labeling 502.3.3 Intention Recognition 522.4 Summary 52Acknowledgments 543 Computer Vision for MAVs 55Friedrich Fraundorfer3.1 Introduction 553.2 System and Sensors 573.3 Ego-Motion Estimation 583.3.1 State Estimation Using Inertial and Vision Measurements 583.3.2 MAV Pose from Monocular Vision 623.3.3 MAV Pose from Stereo Vision 633.3.4 MAV Pose from Optical Flow Measurements 653.4 3D Mapping 673.5 Autonomous Navigation 713.6 Scene Interpretation 723.7 Concluding Remarks 734 Exploring the Seafloor with Underwater Robots 75Rafael Garcia, Nuno Gracias, Tudor Nicosevici, Ricard Prados, Natalia Hurtos, Ricard Campos, Javier Escartin, Armagan Elibol, Ramon Hegedus and Laszlo Neumann4.1 Introduction 754.2 Challenges of Underwater Imaging 774.3 Online Computer Vision Techniques 794.3.1 Dehazing 794.3.2 Visual Odometry 844.3.3 SLAM 874.3.4 Laser Scanning 914.4 Acoustic Imaging Techniques 924.4.1 Image Formation 924.4.2 Online Techniques for Acoustic Processing 954.5 Concluding Remarks 98Acknowledgments 995 Vision-Based Advanced Driver Assistance Systems 100David Gerónimo, David Vázquez and Arturo de la Escalera5.1 Introduction 1005.2 Forward Assistance 1015.2.1 Adaptive Cruise Control (ACC) and Forward Collision Avoidance (FCA) 1015.2.2 Traffic Sign Recognition (TSR) 1035.2.3 Traffic Jam Assist (TJA) 1055.2.4 Vulnerable Road User Protection 1065.2.5 Intelligent Headlamp Control 1095.2.6 Enhanced Night Vision (Dynamic Light Spot) 1105.2.7 Intelligent Active Suspension 1115.3 Lateral Assistance 1125.3.1 Lane Departure Warning (LDW) and Lane Keeping System (LKS) 1125.3.2 Lane Change Assistance (LCA) 1155.3.3 Parking Assistance 1165.4 Inside Assistance 1175.4.1 Driver Monitoring and Drowsiness Detection 1175.5 Conclusions and Future Challenges 1195.5.1 Robustness 1195.5.2 Cost 121Acknowledgments 1216 Application Challenges from a Bird’s-Eye View 122Davide Scaramuzza6.1 Introduction to Micro Aerial Vehicles (MAVs) 1226.1.1 Micro Aerial Vehicles (MAVs) 1226.1.2 Rotorcraft MAVs 1236.2 GPS-Denied Navigation 1246.2.1 Autonomous Navigation with Range Sensors 1246.2.2 Autonomous Navigation with Vision Sensors 1256.2.3 SFLY: Swarm of Micro Flying Robots 1266.2.4 SVO, a Visual-Odometry Algorithm for MAVs 1266.3 Applications and Challenges 1276.3.1 Applications 1276.3.2 Safety and Robustness 1286.4 Conclusions 1327 Application Challenges of Underwater Vision 133Nuno Gracias, Rafael Garcia, Ricard Campos, Natalia Hurtos, Ricard Prados, ASM Shihavuddin, Tudor Nicosevici, Armagan Elibol, Laszlo Neumann and Javier Escartin7.1 Introduction 1337.2 Offline Computer Vision Techniques for Underwater Mapping and Inspection 1347.2.1 2D Mosaicing 1347.2.2 2.5D Mapping 1447.2.3 3D Mapping 1467.2.4 Machine Learning for Seafloor Classification 1547.3 Acoustic Mapping Techniques 1577.4 Concluding Remarks 1598 Closing Notes 161Antonio M. LópezReferences 164Index 195