<div dir="ltr"><div dir="ltr" style="box-sizing:border-box;color:rgb(44,54,58);font-family:Roboto,sans-serif;font-size:14px"><span style="box-sizing:border-box;font-weight:bolder;font-family:Calibri,sans-serif;font-size:11pt"><span style="box-sizing:border-box;font-size:12pt;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial">ROAD @ ICCV 2021: Final Call for Papers</span></span><br style="box-sizing:border-box"></div><div dir="ltr" style="box-sizing:border-box;color:rgb(44,54,58);font-family:Roboto,sans-serif;font-size:14px"><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"> </span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial">The deadline for submitting papers to the ICCV 2021 workshop</span><span style="box-sizing:border-box;font-size:12pt"><br style="box-sizing:border-box"><br style="box-sizing:border-box"><span style="box-sizing:border-box;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial">"The ROAD challenge: Event detection for situation awareness in autonomous driving",</span><br style="box-sizing:border-box"><br style="box-sizing:border-box"><span style="box-sizing:border-box;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial">has been extended to ***August 5 2021***</span><br style="box-sizing:border-box"><br style="box-sizing:border-box"></span><span style="box-sizing:border-box;font-size:12pt"><a href="https://urldefense.com/v3/__https://lnkd.in/gsBU5wU__;!!LIr3w8kk_Xxm!7HKLpdUtkcNkdHeVMOcfLKJGEMdBFxuMRi03k_7AXSGbCfv4I-vcJ5AIbtQA0lhMpidx1PB8$" rel="noreferrer" target="_blank" style="color:blue;box-sizing:border-box;text-decoration-line:none;background-color:transparent"><span style="box-sizing:border-box;color:rgb(17,85,204);background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial">https://lnkd.in/gsBU5wU</span></a></span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt"> </span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt">Decisions will be made public on August 12. Camera-ready will need to be submitted by August 17 to fit the ICCV 2021 publication timeline.</span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt"> </span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt">The deadline for submitting entries to the Challenge has also been extended to ***September 30 2021***</span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt"> </span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt"><a href="https://urldefense.com/v3/__https://sites.google.com/view/roadchallangeiccv2021/challenge__;!!LIr3w8kk_Xxm!7HKLpdUtkcNkdHeVMOcfLKJGEMdBFxuMRi03k_7AXSGbCfv4I-vcJ5AIbtQA0lhMpstPalae$" rel="noreferrer" target="_blank" style="color:blue;box-sizing:border-box;text-decoration-line:none;background-color:transparent">https://sites.google.com/view/roadchallangeiccv2021/challenge</a><br style="box-sizing:border-box"><br style="box-sizing:border-box"><span style="box-sizing:border-box;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial">The goal of this Workshop/Challenge is to put to the forefront of the research in autonomous driving "situation awareness", intended as the ability to create semantically useful representations of dynamic road scenes in terms of the notion of 'road event', leveraging our recently released ROad event Awareness Dataset for autonomous driving (ROAD):</span><br style="box-sizing:border-box"><br style="box-sizing:border-box"></span><span style="box-sizing:border-box;font-size:12pt"><a href="https://urldefense.com/v3/__https://lnkd.in/gjr9M5Z__;!!LIr3w8kk_Xxm!7HKLpdUtkcNkdHeVMOcfLKJGEMdBFxuMRi03k_7AXSGbCfv4I-vcJ5AIbtQA0lhMps29vh55$" rel="noreferrer" target="_blank" style="color:blue;box-sizing:border-box;text-decoration-line:none;background-color:transparent"><span style="box-sizing:border-box;color:rgb(17,85,204);background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial">https://lnkd.in/gjr9M5Z</span></a></span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt"> </span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt">The Challenge proposes three video-level detection tasks: (1) agent detection, (2) action detection and (3) road event detection, intended as a triplet (agent, action, location).</span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt">All tasks consist in regressing 'tubes' of temporally linked bounding boxes and the associated label(s). Researchers active in object detection or action detection are especially encouraged to participate!</span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt"> </span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt">The dataset is composed by 22 long duration (8-minute-long) videos in diverse weather conditions, making the above detection tasks extremely challenging.</span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt"> </span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt">As for the papers, we invite contributions on the following topics:<br style="box-sizing:border-box"><br style="box-sizing:border-box">Detecting and modelling 'atomic' events.</span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt">Detecting and modelling complex activities.</span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt">Predicting agent intentions.<br style="box-sizing:border-box">Dynamic scene understanding from streaming videos.<br style="box-sizing:border-box">Predicting the trajectory of pedestrians, vehicles and other road users.<br style="box-sizing:border-box">Forecasting future road events (both atomic and complex).<br style="box-sizing:border-box">Decision making, both via reinforcement/imitation learning and via intermediate representations, and a critical/empirical comparison between the two approaches.<br style="box-sizing:border-box">Explicability of both perception and decision making components of autonomous driving.<br style="box-sizing:border-box">Modelling road scenarios in a multi-agent framework.<br style="box-sizing:border-box">Modelling the reasoning processes of road agents in terms of goals or mental states.<br style="box-sizing:border-box">Machine theory of mind for autonomous vehicles.<br style="box-sizing:border-box">The role of incremental, life-long and continual learning in autonomous driving, with a focus on situation awareness.<br style="box-sizing:border-box">The use of realistic simulations to generate training data for semantic scene understanding.<br style="box-sizing:border-box">Testing and certification of AI algorithms for autonomous driving.<br style="box-sizing:border-box">The ethical implications of situation awareness and automated decision making.</span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt"> </span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt">Papers will be published as posters in two sessions. The winners of a Best Paper and a Best Student Paper award will instead give an oral presentation, as will the authors of the top entry in each Challenge task.<br style="box-sizing:border-box"><br style="box-sizing:border-box">Papers must follow the ICCV 2021 template and be submitted via CMT:<br style="box-sizing:border-box"><br style="box-sizing:border-box"><a href="https://urldefense.com/v3/__https://lnkd.in/gweCKpU__;!!LIr3w8kk_Xxm!7HKLpdUtkcNkdHeVMOcfLKJGEMdBFxuMRi03k_7AXSGbCfv4I-vcJ5AIbtQA0lhMpgzAKysy$" rel="noreferrer" target="_blank" style="color:blue;box-sizing:border-box;text-decoration-line:none;background-color:transparent"><span style="box-sizing:border-box;color:rgb(17,85,204)">https://lnkd.in/gweCKpU</span></a><br style="box-sizing:border-box"><br style="box-sizing:border-box">The Workshop will allow for the submission of papers concurrently submitted elsewhere to aggregate all relevant efforts in this area.<br style="box-sizing:border-box"><br style="box-sizing:border-box">We have secured an array of prestigious Invited Speakers:<br style="box-sizing:border-box"><br style="box-sizing:border-box">Raquel Urtasun (Toronto)<br style="box-sizing:border-box">Adrien Gaidon (TRI)<br style="box-sizing:border-box">Daniela Rus (MIT)<br style="box-sizing:border-box">Deva Ramanan (CMU)<br style="box-sizing:border-box">Paul Newman (Oxford  University)</span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt"> </span></p><p style="box-sizing:border-box;margin:0cm 0cm 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial;font-size:11pt;font-family:Calibri,sans-serif"><span style="box-sizing:border-box;font-size:12pt">The workshop will also host a panel discussion on the future and ethical implications of autonomous driving.<br style="box-sizing:border-box"><br style="box-sizing:border-box">Organisers<br style="box-sizing:border-box"><br style="box-sizing:border-box">Fabio Cuzzolin, Andrew Bradley (Oxford Brookes University)<br style="box-sizing:border-box">Gurkirt singh (ETH)<br style="box-sizing:border-box">Reza Javanmard Alitappeh (Mazandaran University)<br style="box-sizing:border-box">Stanislao Grazioso, Giuseppe Di Gironimo, Valentina Fontana (Federico II  University, Naples)<br style="box-sizing:border-box">Valentina Mușat (Oxford  University)</span></p></div></div>