Namitha Padmanabhan retweeted
Our solution? Just 4 conv blocks, trained for 5% more epochs.
ConvRefiner restores much of the missing high-frequency detail while keeping the diffusion transformer frozen.
And here's a surprising result: FID barely changes, even when the recovered detail is clearly visible
Namitha Padmanabhan retweeted
What if we let training decide what the model should predict?
MFlow learns whether to predict clean images (x₀), velocity (u), or something in between -- independently along each PCA direction.
No expensive parameterization search. Just learn the choice jointly with the model.
Replying to @VirginAtlantic
@VirginAtlantic It's been 4 days since my BLR→BOM→LHR→IAD trip and there's still no sign of my bags. BOM→LHR leg was delayed and I missed my connection. Was rebooked onto @united, arriving 19 hours late. But my bags never made it to IAD! @united says they never scanned them.
After several emails, @VirginAtlantic admits tags last scanned at BLR by partner @IndiGo6E, but refuses to trace them. Filed a claim with @united who are responsible as the last carrier but they haven't yet heard back from Virgin. Bags still stuck somewhere between BLR & nowhere!
2
5
155
I was also quietly downgraded from my paid Economy Delight fare class to standard economy on United with zero notice or compensation. Pitiful customer service from @VirginAtlantic overall! #LostBaggage #CustomerExperience #flightdelay
1
5
119
Namitha Padmanabhan retweeted
🎉 Excited to share our paper "Trokens: Semantic-Aware Relational Trajectory Tokens for Few-Shot Action Recognition" has been accepted to #ICCV2025!
Equally co-led with @ShuaiyiH — we advance few-shot action recognition via smart point tracking.
🔗 trokens-iccv25.github.io
🧵👇
Namitha Padmanabhan retweeted
We introduce LEIA, a novel approach for representing 3D articulated objects by learning view-invariant latent embeddings for different articulation states. Visit our poster #299 at @eccvconf #ECCV2024 in Milan on Oct 3rd (Thursday), 10:30am.
Project Page: archana1998.github.io/leia/
Namitha Padmanabhan retweeted
📢 Point tracking 🤝 action recognition at #ECCV2024
We've set the new SoTA of few-shot action recognition by harnessing morion data from point tracking and semantic features from SSL.
Curious? Visit Poster #203 Thursday AM to see the future of action recognition🔥. Details:🧵
Namitha Padmanabhan retweeted
Unconditional #diffusion pretraining learns discriminative representation; hence are great candidates for Unified Models for #GenAI as well as #recognition.
Please join us at our #DiffSSL poster (#101) at #ECCV on 1 Oct 1030AM.
@eccvconf – at Milano, Lombardia
"Do text-free diffusion models learn discriminative visual representations?"
tl;dr:
Yes, diffusion models learn highly useful, discriminative visual representations even w/o text! #DiffSSL
website:mgwillia.github.io/diffssl
arxiv:arxiv.org/abs/2311.17921
@zhoutianyi @abhi2610🧵
I will be speaking about my work in INRs (namithap10.github.io/xinc/) at the INRV workshop at #CVPR2024. Come see me in Summit 335-336 at 5PM on Tuesday!







