More Like this

1. Hindsight is 20/20: Leveraging Past Traversals to Aid 3D Perception

   You, Yurong; Luo, Katie Z; Chen Xiangyu; Chen, Junan; Chao, Wei-Lun; Sun, Wen; Hariharan, Bharath; Campbell, Mark; Weinberger, Kilian Q. (April 2022, International Conference on Learning Representations)

   Self-driving cars must detect vehicles, pedestrians, and other traffic participants accurately to operate safely. Small, far-away, or highly occluded objects are particularly challenging because there is limited information in the LiDAR point clouds for detecting them. To address this challenge, we leverage valuable information from the past: in particular, data collected in past traversals of the same scene. We posit that these past data, which are typically discarded, provide rich contextual information for disambiguating the above-mentioned challenging cases. To this end, we propose a novel end-to-end trainable Hindsight framework to extract this contextual information from past traversals and store it in an easy-to-query data structure, which can then be leveraged to aid future 3D object detection of the same scene. We show that this framework is compatible with most modern 3D detection architectures and can substantially improve their average precision on multiple autonomous driving datasets, most notably by more than 300% on the challenging cases. Our code is available at https://github.com/YurongYou/Hindsight. 

   more » « less
2. Hindsight is 20/20: Leveraging Past Traversals to Aid 3D Perception

   Yurong, You; Luo, Katie Z.; Chen, Xiangyu; Chen, Junan; Chao, Wei-Lun; Sun, Wen; Hariharan, Bharath; Campbell, Mark; Weinberger, Kilian Q. (June 2022, International Conference on Learning Representations (ICLR))
3. Hindsight is 20/20: Leveraging Past Traversals to Aid 3D Perception

   You, Yurong; Luo, Katie; Chen, Xiangyu; Chen, Junan; Chao, Wei-Lun; Sun, Wen; Hariharan, Bharath; Campbell, Mark; Weinberger, Kilian Q (April 2022, International Conference on Learning Representations)
4. Introduction: 20 Years of ISEE

   https://doi.org/10.58021/S5V883  

   Seagroves, Scott; Barnes, Austin; Metevier, Anne; Porter, Jason; Hunter, Lisa (January 2022, Institute for Scientist & Engineer Educators (ISEE))

   Seagroves, Scott; Barnes, Austin; Metevier, Anne; Porter, Jason; Hunter, Lisa (Ed.)

   {} 

   more » « less
5. Probing gravity by holding atoms for 20 seconds

   https://doi.org/10.1126/science.aay6428  

   Xu, Victoria; Jaffe, Matt; Panda, Cristian D.; Kristensen, Sofus L.; Clark, Logan W.; Müller, Holger (November 2019, Science)

   null (Ed.)

   Atom interferometers are powerful tools for both measurements in fundamental physics and inertial sensing applications. Their performance, however, has been limited by the available interrogation time of freely falling atoms in a gravitational field. By suspending the spatially separated atomic wave packets in a lattice formed by the mode of an optical cavity, we realize an interrogation time of 20 seconds. Our approach allows gravitational potentials to be measured by holding, rather than dropping, atoms. After seconds of hold time, gravitational potential energy differences from as little as micrometers of vertical separation generate megaradians of interferometer phase. This trapped geometry suppresses the phase variance due to vibrations by three to four orders of magnitude, overcoming the dominant noise source in atom-interferometric gravimeters. 

   more » « less