bharat_b7

Research Scientist, Meta

[Nikos Athanasiou @_nikos_athan]

🧨 RHOBIN @CVPR announcement 🎇
We are pleased to receive the generous sponsorships of @meshcapade and @GoogleDeepMind 🤗🤗
Given their amazing support, we will be distributing ~4k among the winners of our challenges!
We are waiting for your submissions (DDL: 30/05) 🥳
@CVPRConf

[Nikos Athanasiou @_nikos_athan]

📢 Call for participants: 3rd challenge on reconstruction of humans, objects, and interactions. 

Some previous winners have published their winning methods at top-conferences, and there will be rewards from our sponsors🤗🤗 
@CVPR
@CVPRConf
#CVPR #CVPR25

[Xianghui Xie @XianghuiXie]

📢#CVPR workshop call for papers. Are you working on:
- 3D humans
- 3D objects
- interactions for humans & objects
Welcome to submit to our 3rd RHOBIN workshop! We have an amazing line of speakers! @CVPRConf @CVPRConf
details see: https://t.co/XxvfkNMZdn

[Nikos Athanasiou @_nikos_athan]

That is what happens if you win RHOBIN challenge at CVPR 😉
You get accepted to ECCV :)
Congrats to the authors for their nice work and big thanks for choosing our challenge as a benchmark!

[Nikos Athanasiou @_nikos_athan]

RHOBIN is not to be missed, fantastic up of speakers! @Michael_J_Black always gives great talks!

[Siwei Zhang @SiweiZhang13]

Check our work at CVPR! A collaboration effort with @bharat_b7 @ Yuanlu Xu @awinkler_ @ Petr Kadlecek @SiyuTang3 @FedericaBogo

📍Oral 4C, Summit Flex Hall C
🗓️1.00-2.30pm, Thursday
📍Poster 182 @ poster session 4, Arch 4A-E
🗓️5.15-6.45pm, Thursday

[Dimitris Tzionas @dimtzionas]

Do you work on 3D Human-Object Interaction? 📢 You still have 1 week to submit to our #CVPR24 @CVPR challenge - ddl May 30🆘 📢 $3000 will be split to winners thanks to @meshcapade, @SpreeAI, @Apple. Challenge co-organizers: @XianghuiXie, @sha2nk_t , @bharat_b7, @GerardPonsMoll1

[Xianghui Xie @XianghuiXie]

Reconstruct human interacting in real world.
Deadline: May30.
Still one more week to submit your results!
Join us soon!

[Nikos Athanasiou @_nikos_athan]

RHOBIN challenge with awards!@CVPR
Deadline: May 30th!
3D humans interacting with the world✌️
Extra motivation: One of my most cited papers comes from my undergrad, out of a challenge in @SemEvalWorkshop
Main conferences are nice, but you never know where it starts from! :)

[Xi @xiwang1212]

@CVPR We have an amazing lineup of speakers Carl Vondrick @cvondrick, Dima Damen @dimadamen, David Fouhey, Angela Dai @angelaqdai, Michale Black @Michael_J_Black

[Xi @xiwang1212]

The second Rhobin Challenge is coming to #CVPR24 @CVPR!

Join the interaction modeling community by contributing a paper or participating in the challenges, with the opportunity to win awards!

[Gerard Pons-Moll @GerardPonsMoll1]

work by @bernakabadayi Wocjciek Zielonka @bharat_b7 and @JustusThies and myself.

Project: https://t.co/zhEpJqnEQK
Arxiv: https://t.co/4TLnPTk3yp

Could be a nice read for your trip home for Christmas :-)

[Gerard Pons-Moll @GerardPonsMoll1]

Our virtual humans lecture videos on YouTube:
https://t.co/rJCb6JaabA

We cover from the basics (image formation, surface reps), to classical models like SMPL to modern NeRF based ones, as well as models of human motion in scenes based on transformers, diffusion, etc. Enjoy!

[Yuxuan Xue @yxue_yxue]

How to combine the implicit surface and classical mesh representation? Define the neural field solely on the surface! #ICCV2023

We present NSF: Neural Surface Fields for Human Modelling from Monocular Depth on Thursday 2:30-4:30 at Room "Foyer Sud" - 100. See you there!

[Gerard Pons-Moll @GerardPonsMoll1]

Do you want to know how to track humans and objects through occlusions, and get motion in a global coordinate frame?

Come to our poster 59 west building.
#CVPR2023
Xianghui xie
@bharat_b7

[bharat_b7 @bharat_b7]

Track #HumanObjectInteractions from a video in 3D. Transformer based VisTracker reasons about occlusions and leverages surrounding frames to make correct predictions under occlusion. https://t.co/15FzEqjE8l
Poster: Tuesday, 1630PM, West Exhibit Nr. 59 #CVPR23 #3DHumans

[bharat_b7 @bharat_b7]

Now you can synthesise human-object interaction motion with explicit control on contacts!
Hall B, poster #6, 15:30
#eccv2022 @GerardPonsMoll1 @blacksquirrel__ @Xiaohan23543137

[Xiaohan Zhang @xiaothedon]

We are presenting COUCH, the dataset and the method of human-chair interactions at #ECCV2022. Drop by our poster (#6) in Hall B this afternoon between 15:30 and 17:30 to have a quick chat. Look forward!

project page: https://t.co/zIL2F7DQNo

[bharat_b7 @bharat_b7]

Hate noisy hand trackers? We present TOCH: Spatio-Temporal Object-to-Hand Correspondence for Motion Refinement. Agnostic to the tracker, TOCH uses hand-object interaction prior to correct the interaction.

1530 Poster 1.B hall B, 52
https://t.co/KNQmyHUuCs

@GerardPonsMoll1

[Gerard Pons-Moll @GerardPonsMoll1]

The key is to predict @neural_fields for the human and object surface and a correspondence and pose field to make optimization well behaved.
We also have a nice simple trick to train pixel-aligned reconstructions from perspective cameras.

@XianghuiXie @bharat_b7 and myself

[Julian Chibane @ChibaneJulian]

Thrilled to receive the best paper honorable mention at 3DV'22 - together with Yongqin & @bharat_b7.

Thanks to supervisors @zeynepakata, Bernt & @GerardPonsMoll1!

Our method improves single image 3D reconstruction and generalizes to novel object classes.
https://t.co/5tiv3xmqXo