BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Learning Multi-Scene Absolute Pose Regression with Transformers - 
 Yoli Shavit\, Huawei TRC and Bar-Ilan University\, Israel
DTSTART:20221122T130000Z
DTEND:20221122T140000Z
UID:TALK183470@talks.cam.ac.uk
CONTACT:Pietro Lio
DESCRIPTION:"Join us on Zoom":https://zoom.us/j/99166955895?pwd=SzI0M3pMVE
 kvNmw3Q0dqNDVRalZvdz09\n\nAbsolute camera pose regressors estimate the pos
 ition and orientation of a camera from the captured image alone. Typically
 \, a convolutional backbone with a multi-layer perceptron head is trained 
 with images and pose labels to embed a single reference scene at a time. R
 ecently\, this scheme was extended for learning multiple scenes by replaci
 ng the MLP head with a set of fully connected layers. In this work\, we pr
 opose to learn multi-scene absolute camera pose regression with Transforme
 rs\, where encoders are used to aggregate activation maps with self-attent
 ion and decoders transform latent features and scenes encoding into candid
 ate pose predictions. This mechanism allows our model to focus on general 
 features that are informative for localization while embedding multiple sc
 enes in parallel. We evaluate our method on commonly benchmarked indoor an
 d outdoor datasets and show that it surpasses both multi-scene and state-o
 f-the-art single-scene absolute pose regressors. We make our code publicly
  available from: "https://github.com/yolish/multi-scene-pose-transformer":
 https://github.com/yolish/multi-scene-pose-transformer \n\n*BIO:*\n\nDr Yo
 li Shavit is a Principal Research Scientist Manager at Huawei Tel Aviv Res
 earch Center (TRC) and a Postdoctoral Researcher at Bar-Ilan University. B
 efore joining Huawei\, Yoli worked at Amazon and interned at Microsoft Res
 earch. She holds a PhD in Computer Science from the University of Cambridg
 e\, an MSc in Bioinformatics from Imperial College London and a BSc in Com
 puter Science and in Life Science from Tel Aviv University. Yoli is the re
 cipient of the Cambridge International Scholarship\, and her thesis was no
 minated for the best thesis award in the UK. Her current research focuses 
 on deep learning methods for camera localization and multi-view stereo\, w
 ith recent publications in CVPR\, ICCV\, ECCV and NeurIPS. 
LOCATION:Zoom
END:VEVENT
END:VCALENDAR