Gaurav Shrivastava

Applied Scientist @Amazon | CS PhD @UMDCS | Ex-Research @Google | @Meta | @tiktok_us | @NUSingapore | @bitspilaniindia

[Gaurav Shrivastava @datacrunch3r]

I am pleased to announce that our work Video Decomposition Prior(VDP) have been accepted for ICLR 2024.

Key Highlights:

(1/3)

[devanshi kotak @DevanshiKotak]

Honored to be featured in @Forbes! Tips for first-time PMs on credibility, customer focus, and balancing priorities.
https://t.co/YPcLFJa6En

#ProductManagement #Forbes #CareerGrowth #WomenInTech

[devanshi kotak @DevanshiKotak]

Thrilled to see @Cisco’s IPFM at work during #Paris2024 Olympics

@Francetele delivered seamless UHD broadcasts at scale. Honored to have helped bring this story to life on a global stage.

👉 https://t.co/GyQV3ZvXrZ

#Cisco #IPFM #Olympics #BroadcastTech #MediaInnovation

[Gaurav Shrivastava @datacrunch3r]

@gowthami_s Congratulations Dr. Somepalli 🎉🥳

[devanshi kotak @DevanshiKotak]

I'm thrilled to share that I will be presenting for the first time at #CiscoLive! Join me for an interactive breakout where we'll dive into how microsegmentation can be leveraged in your data center.

Session: IBODCN-1011
https://t.co/xYOvjFPKOa

#Microsegmentation #VXLAN @Cisco

[Harsh Shrivastava @Ace_Harsh]

Thank you @MSFTResearch for highlighting our work on "Generative Kaleidoscopic Networks"! I was trying to create a dataset kaleidoscope using neural networks. In this process, I additionally discovered that neural networks demonstrate an "over-generalization" phenomenon. (1/2)

[Gaurav Shrivastava @datacrunch3r]

OpenReview pdf link: https://t.co/v3jbzO2Jrq

This work is a collaborative effort by myself, @sernamlim and @abhi2610.
Stay tuned for the camera ready version.
#ICLR2024 #videoediting #computationalphotography

(3/3)

[Gaurav Shrivastava @datacrunch3r]

I am pleased to announce that our work Video Decomposition Prior(VDP) have been accepted for ICLR 2024.

Key Highlights:

(1/3)

[Gaurav Shrivastava @datacrunch3r]

1️⃣ The technique showcases proficiency in three critical tasks: video relighting, dehazing, and unsupervised object segmentation
2️⃣ No need for data annotations!
3️⃣ An efficient test-time optimization technique, it demonstrates remarkable performance using just a single GPU

2/3

[Harsh Shrivastava @Ace_Harsh]

Thank you @MSFTResearch for sharing this work! Kindly reach out to me (or my co-author) in case you want to know more or explore potential applications! Thanks :)

[Harsh Shrivastava @Ace_Harsh]

Excited to release our framework for doing multivariate time series segmentation in O(N). It is a cross domain approach based on using sparse graph recovery methods. Do check it out!