It’s a special one as it’s my first time contributing to one. It has been such a fantastic ride 💚
What if we could parse egocentric procedures online, with no training? 🤔
Our work VidParse at #ECCV ’26 ,led by @learn2phoenix, combines manipulation-aware features with graph-constrained decoding for robust online parsing of complex, multi-step activities. 🔥
Check it out! 👇
🧵 VidParse has been accepted to #ECCV2026! 🎉
Can we parse a how-to video as it unfolds - without training?
Yes. And it beats the best online supervised baselines, while coming close to offline performance. 🚀
Paper 👇
arxiv.org/abs/2608.27562
12
470
Pulkit retweeted
🧵 VidParse has been accepted to #ECCV2026! 🎉
Can we parse a how-to video as it unfolds - without training?
Yes. And it beats the best online supervised baselines, while coming close to offline performance. 🚀
Paper 👇
arxiv.org/abs/2608.27562
Pulkit retweeted
Looking forward to presenting Scale Space Diffusion at #CVPR2026!
📍Poster 675 - Poster Session 5, Exhibit Hall F
📅 June 7 (Sunday)
🕒 11:45 AM - 1:45 PM MDT
Joint work with @soumikkanad* and @abhi2610.
Project webpage: prateksha.github.io/projects…
Pulkit retweeted
We’re looking forward to presenting UPLiFT at #CVPR2026! Efficiently extract pixel-dense features from pretrained backbones like DINOv3.
We’ll be at the final poster session on Sunday (6/7) from 3:30-5:30pm at Poster 474, so please come by!
Website: cs.umd.edu/~mwalmer/uplift/
Pulkit retweeted
Come visit our poster "Scale Space Diffusion" at #CVPR2026!
We reformulate diffusion using scale space theory to enable progressive information flow for efficient pixel diffusion!
📍Main Conference: ExHall F 675; Sun, Jun 7, 2026 • 11:45 AM – 1:45 PM MDT
Pulkit retweeted
Excited to announce my first preprint in LM interpretability!
Latent reasoning models are not monitorable by default, since they don't reason in human-readable, natural language text. But can we make progress in understanding their intermediate reasoning steps using mech interp?
Pulkit retweeted
Diffusion models be like:
“this image is 97% noise… better process all 256×256 pixels anyway”
If very noisy diffusion states contain no more useful information than a tiny downsampled image,
Then why run expensive full-res computation on them?
🧵
Pulkit retweeted
Excited to announce that UPLiFT has been accepted to #CVPR2026!
You can also try out UPLiFT right now to extract pixel-dense DINOv3 features with our pretrained models linked below!
Code: github.com/mwalmer-umd/UPLiF…
Paper: arxiv.org/abs/2601.17950
Website: cs.umd.edu/~mwalmer/uplift/
Pulkit retweeted
We’re excited to announce UPLiFT, our lightweight, pixel-dense feature upsampler. UPLiFT boosts feature density, preserves semantics, and has better efficiency scaling than recent SOTA methods. See all links in the thread below.
Coauthors: @_sakshams_ @AnirudAgg @abhi2610
🧵[1/6]
Pulkit retweeted
🚨 PhD Opening: 3D/4D SLAM & Inverse Rendering for Endoscopy (NIH funded) at @UNC_CS! 🚨
The Project:
🟩 Ideal for students with a strong background in 3D Computer Vision & interested in medical robotics & visualization.
🟩 Highly collaborative work with roboticists, medical imaging researchers, and clinicians (GI, ENT, Pulmonology).
🟩Aim to publish in top-tier CV/ML/Robotics/Medical Imaging venues.
🟩 Build on our prior works:
1⃣Endoscopy depth estimator: PPSNet [ECCV'24 - ppsnet.github.io/]
2⃣ 3D SLAM for endoscopy: NFL-BA [NeurIPS'25 -asdunnbe.github.io/NFL-BA/]
Apply now! 📧 Email CV to [email protected] 📝 Formal UNC CS application required (Deadline soon!)
RTs appreciated! 🙏
Aloha #ICCV2025! 🌺
Join @ShuaiyiH, @abhi2610 and me today (21 Oct) at poster #334, 11 AM–1 PM!
We will be presenting Trokens — our work on point tracking for action recognition.
Details of the paper are below 👇
🎉 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
🧵👇
4
297
Pulkit retweeted
Everyone says they want general-purpose robots.
We actually mean it — and we’ll make it weird, creative, and fun along the way 😎
Recruiting PhD students to work on Computer Vision and Robotics @umdcs for Fall 2026 in the beautiful city of Washington DC!
🎉 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
🧵👇
👥 This work was co-led with @ShuaiyiH in collaboration with @MatthewWalmer, @rssaketh , and @abhi2610
🔗 All code and data are released:
🌐 Webpage: trokens-iccv25.github.io
📄 ArXiv: arxiv.org/abs/2508.03695
💻 Code: github.com/pulkitkumar95/tro…
🤗 Data: huggingface.co/datasets/pulk…
1
5
208






















