←
Jason Corso1,315 posts
Jason Corso avatar

Jason Corso

@_JasonCorso_

Corso is a Professor at U Michigan and Co-Founder of Voxel51 who makes the category-defining data+model codevelopment ML Tool: FiftyOne

◷ Joined June 2011▧ 147 Photos and videos
Jason Corso avatar
Jason Corso@_JasonCorso_·Nov 13

Could AI be making me appreciate live music more? AI can generate a song. This amazing jazz band made many up together, on the fly, on a brewery patio. 🎷

Post attachment
Jason Corso avatar
Jason Corso@_JasonCorso_·Nov 11

Lancet study: thelancet.com/journals/landi… 698 rural hospitals at risk of closing: beckershospitalreview.com/fi… What we're doing about it in VIGIL: news.engin.umich.edu/2025/02…

Jason Corso avatar
Jason Corso@_JasonCorso_·Nov 11

AI can now read a heart MRI in 2--3 seconds, slightly better than experienced clinicians. Cool. Now find an MRI in rural America. Reading the scan was never the bottleneck. Getting one is.

Post attachment
Jason Corso avatar
Jason Corso@_JasonCorso_·Nov 6

The most impressive model I’ve seen all year had zero hyperparameters. 🌲 Could AI be making me appreciate nature more? We spend months tuning hyperparameters and babysitting training runs to get a model to do one thing well. This forest came from evolution and nobody tweaked anything. It just is. 🌲 (This is a real photo taken recently.)

Post attachment
Jason Corso avatar
Jason Corso@_JasonCorso_·Nov 3

Took my 9yo daughter to vote with me and we read the whole ballot together. Grateful for the right to vote. Go exercise yours!

Jason Corso avatar
Jason Corso@_JasonCorso_·Oct 15

VOXEL51 IS REBORN TODAY!!!! Starting today, anyone can try Voxel51's paid features for FREE with just an email. For anyone who hasn’t tried our product yet, here are three things I’d try first:  → Voxel51 Search: Hybrid search across embeddings and structured metadata. All data uploaded to Voxel51 is automatically indexed, so you can search by semantic similarity alongside exact conditions, metadata, and multimodal events out of the box. → Agentic Labeling: Powered by VLMs, generate labels from natural-language instructions and visual examples, then review and refine the prompts before applying them more broadly. → Voxel51 Agent: Everything you can do with Voxel51 can now be initiated with a prompt. Ask and answer questions about your data and models, and turn those actions into repeatable workflows. Try it on your own data and tell me in the comments what works and what doesn’t: lnkd.in/gx4_kV_C

Post attachment
Jason Corso avatar
Jason Corso@_JasonCorso_·Sep 16

Folks is @ylecun's Keynote from @eccvconf available on YouTube or something? It was probably the most sensitive argument for why physical challenges in AI are different than text-alone. And I have a bunch of Controls friends who need to see it.

Jason Corso avatar
Jason Corso@_JasonCorso_·Sep 16

Farewell ECCV 2026! 🇸🇪🇸🇪🇸🇪🇸🇪 Among the conferences I've been to in the last few years, this ECCV in Malmö was among the most enjoyable. The setting was grand. The size was just right (for these days). Many solid papers....world models and Physical AI ahoy folks! The keynotes by Kristen Grauman, @ylecun Yann LeCun, and @Jamie_Shotton were seriously enjoyable. Enjoyed presented a handful of papers from @Voxel51 and @UMich I left feeling rejuvenated. In a field where it is easy to confuse "paper progress" with real progress, this is just what I needed.

Post attachment
Jason Corso avatar
Jason Corso@_JasonCorso_·Sep 2

From HOMES to SHAME—a lesson in signal vs. noise. For generations, students have learned the Great Lakes acronym: Huron, Ontario, Michigan, Erie, Superior. Now, amidst genuine economic headwinds, shifting immigration policies, and critical challenges facing our workforce and technology sectors, time is being spent trying to rename Lake Ontario to "Lake America." That turns HOMES into SHAME (Superior, Huron, America, Michigan, Erie). As an engineer and founder, my core job is filtering out noise to focus on real-world systems, actionable data, and meaningful human impact. Superficial rebranding doesn't spur innovation, lower mortgage rates, or train the next generation of engineers. It’s pure distraction. Let’s save the energy for problems that actually need solving.

Post attachment
Jimmy at Voxel51 avatar
⟳ Jason Corso retweeted
Jimmy at Voxel51@jimmy_voxel51·Jul 27

FiftyOne now supports multimodal time-series data—visualize, query, and curate synchronized sensor episodes from one platform - hubs.ly/Q04qJrwN0 End-to-end and VLA models generalize better but act as black boxes. When they fail, data is your only window in. Yet most teams still stitch together playback tools, custom scripts, and annotation platforms just to answer one question: is this a one-off, or a pattern? New in FiftyOne: * Native MCAP ingestion—camera, LiDAR, and numeric sensor streams play back in sync, no lossy pre-extraction   * Fully searchable time-series—query by metadata, events, labels, or embedding-based similarity   * Labels persist as tracks across entire episodes   * Temporal tags on any interval, turned into queryable, metadata-rich assets This means root-cause analysis in minutes instead of days. #PhysicalAI #Robotics #AutonomousVehicles #ComputerVision #MLOps

Jimmy at Voxel51 avatar
⟳ Jason Corso retweeted
Jimmy at Voxel51@jimmy_voxel51·Jul 23

"Pick the small one, not the big one." Simple instruction. But if the model underneath can't ground it, your robot grabs the wrong item - voxel51.com/blog/kubricount-… We ran NVIDIA's LocateAnything-3B through KubriCount — a synthetic benchmark built to break grounding models — using a curate → annotate → generate → evaluate loop in FiftyOne. * The results on the hardest split: precision 0.33, recall 0.28 * The model finds fewer than a third of the objects it's asked for * 94 scenes, it counted the distractor instead of the target. Prompted for "boat" (6 in frame), it returned 34 — silently counting the vessels next to them Aggregate metrics hide this. You only catch it by looking at the actual boxes. Check out @datascienceharp 's blog #SyntheticData #Robotics #FoundationModels #ComputerVision

Jimmy at Voxel51 avatar
⟳ Jason Corso retweeted
Jimmy at Voxel51@jimmy_voxel51·Jul 22

Raw, multi-sensor robotics data is notoriously messy to work with - docs.voxel51.com/user_guide/… - FiftyOne solves that: * Unified Visuals: Combining multi-camera feeds, depth maps, and point clouds straight from standard MCAP files. * Temporal Tagging: Easily curating time-series data on the fly (like tagging exact gripper states). * Instant Organization: Turn complex sensor logs into clear, actionable datasets in seconds. Clean data curation is the backbone of great AI and robotics models. #Robotics #DataScience #AITraining #Tech #DataVisualization

1:18