Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Volumetric Capture is Hard There is no magic solution. Anyone interested in spatial R&D can get access to some awesome datasets online to test with, which can be especially interesting when processed with 3DGS and played back as 4DGS. However you need a ton of compute power. (multiple 3080,... 3090+, RTX5000+ etc) The Necessity of Multiple Cameras for Volumetric Capture Consider, for instance, this very useful fixed rig dataset comprising 20 camera streams provided by Facebook Research Or this datasets comprising 40 camera streams provided by Google The attached video demonstrates FB's RAW 3DGS data played back in real-time in our custom Sibr viewer. It's noticable how quickly the quality can breakdown out of the camera group frustrum. Even at this quality it takes a lot of time and energy to process and playback. It's not simple. Let alone the expertise required by the original research team who captured the data. Doing this with 20 cameras is hard. 150+ cameras even harder. Now imagine trying to achieve it with a single camera phone to playback on a Meta Quest or Spatial headset. #volumetriccaptureishard #camerasdontseearoundcorners #spatialshow more

Infinite-Realities

8,708 subscribers

129,521 views • 2 years ago •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

10 Comments

3D Scanstore2 years ago

It's surprising how many people would have you believe that cameras can see around corners :) "We reconstructed this entire cathedral from just one image” There’s a lot of selective camera angles and exaggerated claims circulating currently. The work you're doing is setting the bar. I wouldn’t concern yourself with the misleading information out there; ultimately, the quality of your work and the actual complexity of 360 degree 4D video will become apparent quite quickly once clients start asking for it.

Infinite-Realities2 years ago

Thanks that's most appreciated. The missinformation does tend to muddy the waters though, which makes it a little harder to navigate becuase of false expectations.

SMOUSE 🔸2 years ago

The distortion could make a really cool movie effect about looking through time, kinda like interstellar

Alvinoo2 years ago

Something like a 4DGS in Soccer or UFC should be amazing. Preinstalled cameras recording from different angles, pit dimensions set and don’t require rendering. Only 22 players + ref. Could work?

Infinite-Realities2 years ago

It's been done for Football by Canon and other companies in the past. Also it's worth a google for the Intel stage (which shuttered). The results are OK. UFC would be cool for sure. You would still need a ton of cameras. 150-200+ it's a multi-million £ project.

Mitya Shabat2 years ago

Yeah, very hard not to overfit to the provided camera set, even if there's a lot Lots of ideas in research on how to regularize these 3D representations to work with sparse input, but these are mostly heuristics and hard to configure properly. This is hard. I love it

Zvezdan Nedeljkovic2 years ago

It would be useful to separate the subject somehow from the rest of the scene and if nothing changes in the background keep the background the same until something changes. With reflective surfaces it would be much harder of course.

Infinite-Realities2 years ago

That's the research we are currently doing. We were just demonstrating real-world capture results. Capturing in situ is hard.

⚡LaserEyeMaxi⚡2 years ago

Post it on

Tim Reha2 years ago

Streaming holodeck

Related Videos

We're proud to showcase some recent real-time work with 3D / 4D Gaussian splatting. In January of 2024 we acquired a full 3DGS commercial license with Inria to allow us to offer 4DGS services to clients around the world. We've spent the last 10 months developing our custom 4DGS software pipeline, utilizing our own Nvidia GPU cluster to batch process 100,000's of 4DGS frames rapidly, combined with our state-of-the-art custom built volumetric capture system(s). We can capture anything from complex character interactions for visual effects shots, or dynamic fast fighting scenes, sports, acting, even capturing and storing memories of family and friends for the future! The possibilities are endless. Our focus is always on high-quality and high-fidelity. We don't cut corners and we're continually striving to learn and improve. The below video showcases some 6-dof spatial captures we took of Henry Pearce's family. These scenes are played back in real-time @ 30fps inside a 3rd party engine running in a 120fps virtual environment. This is a culmination of years of hard work and dedication and we are keen to commercialize this technology with you. Please do reach out if you're interested in our capture systems or our 3D/4DGS processing services. HQ YouTube Link: HQ Vimeo Link: #spatialcapture #vr #virtualmemories #inria #3dgs #4dgs #aeonx #ir #ximea Music "Vision" by megiddo music Special thanks to Aras Pranckevičius 🇺🇦🇱🇹

We're proud to showcase some recent real-time work with 3D / 4D Gaussian splatting. In January of 2024 we acquired a full 3DGS commercial license with Inria to allow us to offer 4DGS services to clients around the world. We've spent the last 10 months developing our custom 4DGS software pipeline, utilizing our own Nvidia GPU cluster to batch process 100,000's of 4DGS frames rapidly, combined with our state-of-the-art custom built volumetric capture system(s). We can capture anything from complex character interactions for visual effects shots, or dynamic fast fighting scenes, sports, acting, even capturing and storing memories of family and friends for the future! The possibilities are endless. Our focus is always on high-quality and high-fidelity. We don't cut corners and we're continually striving to learn and improve. The below video showcases some 6-dof spatial captures we took of Henry Pearce's family. These scenes are played back in real-time @ 30fps inside a 3rd party engine running in a 120fps virtual environment. This is a culmination of years of hard work and dedication and we are keen to commercialize this technology with you. Please do reach out if you're interested in our capture systems or our 3D/4DGS processing services. HQ YouTube Link: HQ Vimeo Link: #spatialcapture #vr #virtualmemories #inria #3dgs #4dgs #aeonx #ir #ximea Music "Vision" by megiddo music Special thanks to Aras Pranckevičius 🇺🇦🇱🇹

Infinite-Realities

202,783 views • 2 years ago

Americans are under surveillance Roughly 40,000-60,000 cameras have been placed throughout 42 states that are tagging your vehicle and entering it into a database to track you “These cameras capture more than your license. They can capture the interior of your vehicle as you drive by these cameras, especially the positioning of the camera and it's legal (Attorney shows a picture of what one of these “Flock Cameras” looks like, when you see it you’ll instantly recognize it. They’re everywhere) “So what happens is this camera captures your tag and when it captures your tag is able to detect where your vehicle has been anywhere in the United States where there are flock cameras. If you're in Georgia and you travel to New York, every time your vehicle tag gets captured by a flock camera, it is entered into a database and they are able to track it.”

Americans are under surveillance Roughly 40,000-60,000 cameras have been placed throughout 42 states that are tagging your vehicle and entering it into a database to track you “These cameras capture more than your license. They can capture the interior of your vehicle as you drive by these cameras, especially the positioning of the camera and it's legal (Attorney shows a picture of what one of these “Flock Cameras” looks like, when you see it you’ll instantly recognize it. They’re everywhere) “So what happens is this camera captures your tag and when it captures your tag is able to detect where your vehicle has been anywhere in the United States where there are flock cameras. If you're in Georgia and you travel to New York, every time your vehicle tag gets captured by a flock camera, it is entered into a database and they are able to track it.”

Wall Street Apes

329,561 views • 1 year ago

The largest pack in the Greater Voyageurs Ecosystem this past winter: the Blackstone Pack. Nice of them to give the camera a few nibbles on their way by! If you enjoy videos like this, we need your help because almost all of our trail camera work is supported by donors. This footage was literally captured because of donors. We simply need folks who enjoy such content to make a small donation ($10, $25, $50) to our project at the link below to help us cover the costs associated with keeping 350-400 trail cameras going year round. $30 purchases batteries for a camera, $200 purchases a new camera set-up. Think about it: most people regularly pay $10-20 per month for Netflix, Hulu, etc. so those companies can create endless content and make money. All we are asking for is $10-20 a year to help us cover our costs, and keep this work going so we can share it with everyone. Of course, the main point of the cameras is wolf research (and now, a little cougar research on the side) but cool footage of wildlife is a byproduct of having so many cameras out. So help us out and donate here:

The largest pack in the Greater Voyageurs Ecosystem this past winter: the Blackstone Pack. Nice of them to give the camera a few nibbles on their way by! If you enjoy videos like this, we need your help because almost all of our trail camera work is supported by donors. This footage was literally captured because of donors. We simply need folks who enjoy such content to make a small donation ($10, $25, $50) to our project at the link below to help us cover the costs associated with keeping 350-400 trail cameras going year round. $30 purchases batteries for a camera, $200 purchases a new camera set-up. Think about it: most people regularly pay $10-20 per month for Netflix, Hulu, etc. so those companies can create endless content and make money. All we are asking for is $10-20 a year to help us cover our costs, and keep this work going so we can share it with everyone. Of course, the main point of the cameras is wolf research (and now, a little cougar research on the side) but cool footage of wildlife is a byproduct of having so many cameras out. So help us out and donate here:

Voyageurs Wolf Project

12,472 views • 2 months ago

Security Camera App: DVR Recorded Video Playback Watch this video to see how easy it is to remotely playback recorded video surveillance footage on a Viewtron DVR with the mobile app for iPhone and Android. You can do this at the location of your security camera system using a WIFI connection or remotely over the Internet from anywhere in the world. It's easy to check in on your property using a Viewtron camera system when you are traveling. You can learn more about the Viewtron mobile app here.

Security Camera App: DVR Recorded Video Playback Watch this video to see how easy it is to remotely playback recorded video surveillance footage on a Viewtron DVR with the mobile app for iPhone and Android. You can do this at the location of your security camera system using a WIFI connection or remotely over the Internet from anywhere in the world. It's easy to check in on your property using a Viewtron camera system when you are traveling. You can learn more about the Viewtron mobile app here.

CCTV Camera Pros

30,806 views • 1 year ago

Eagle Sports gun range removes Flock cameras over privacy concerns ----- So the cameras were in place since February. Someone didn't make a viral video about it until now (5 - 6 months later). If you can make the time, go to a Flock camera near you. When you get to the camera, if it's near something "sensitive", like a gun range, or a marijuana dispensary, a bar, anything that the public should know about, make a video and post it. Tag me if you like, and/or some of the other flock pages you follow so we can help bring attention to the video. Maybe we can put pressure on the camera owner like we have with this business and get the camera taken down.

Eagle Sports gun range removes Flock cameras over privacy concerns ----- So the cameras were in place since February. Someone didn't make a viral video about it until now (5 - 6 months later). If you can make the time, go to a Flock camera near you. When you get to the camera, if it's near something "sensitive", like a gun range, or a marijuana dispensary, a bar, anything that the public should know about, make a video and post it. Tag me if you like, and/or some of the other flock pages you follow so we can help bring attention to the video. Maybe we can put pressure on the camera owner like we have with this business and get the camera taken down.

Orwell Day

39,667 views • 1 day ago

Everybody is stepping into the LOD game, and so is LichtFeld Studio. Preview of 260M Gaussians streaming into the viewer live. I use the RAD format which is processed on GPU within LichtFeld. It can be also simply dumped straight into the spark.js web viewer, albeit it will die at that amount of Gaussians. What other solutions don't tell you is that they need hours to preprocess a 3DGS ply to make it streamable. This was just a ply exported to RAD by LichtFeld Studio's convert tool (took 5 min at that size) and it is immediately ready to stream. In the comments there is a smaller dataset with 103M Gaussians that streams on startup into the viewer. Both datasets were created by Andrii Shramko. Let's see how far I can push this (need bigger datasets)

Everybody is stepping into the LOD game, and so is LichtFeld Studio. Preview of 260M Gaussians streaming into the viewer live. I use the RAD format which is processed on GPU within LichtFeld. It can be also simply dumped straight into the spark.js web viewer, albeit it will die at that amount of Gaussians. What other solutions don't tell you is that they need hours to preprocess a 3DGS ply to make it streamable. This was just a ply exported to RAD by LichtFeld Studio's convert tool (took 5 min at that size) and it is immediately ready to stream. In the comments there is a smaller dataset with 103M Gaussians that streams on startup into the viewer. Both datasets were created by Andrii Shramko. Let's see how far I can push this (need bigger datasets)

MrNeRF

23,651 views • 2 months ago

Northern Minnesota feels like it is getting a bit wilder. Footage from just a few weeks ago. This does not appear to be the mother of the kittens. Folks tell us all the time how much they enjoy our trail camera videos, and our typical trail camera video is viewed >100,000-200,000 times and liked by 4,000-5,000 people. We simply need folks who enjoy our content to make a small donation ($10, $25, $50) to our project at the link below to help us cover the costs associated with keeping 350-400 trail cameras going year round. Quite literally, if a small percent (2-3%) of folks who enjoy our videos make a small donation to our project at the link in the comments, we reach our fundraising goal and it allows us to keep sharing wild moments like this with everyone for free! Think about it: most people regularly pay $10-20 per month for Netflix, Hulu, etc. so those companies can create endless content and make money. All we are asking for is $10-20 a year to help us cover our costs, and keep this work going so we can share it with everyone. Of course, the main point of the cameras is wolf research (and now, a little cougar research on the side) but cool footage of wildlife is a byproduct of having so many cameras out. So help us out and donate here:

Northern Minnesota feels like it is getting a bit wilder. Footage from just a few weeks ago. This does not appear to be the mother of the kittens. Folks tell us all the time how much they enjoy our trail camera videos, and our typical trail camera video is viewed >100,000-200,000 times and liked by 4,000-5,000 people. We simply need folks who enjoy our content to make a small donation ($10, $25, $50) to our project at the link below to help us cover the costs associated with keeping 350-400 trail cameras going year round. Quite literally, if a small percent (2-3%) of folks who enjoy our videos make a small donation to our project at the link in the comments, we reach our fundraising goal and it allows us to keep sharing wild moments like this with everyone for free! Think about it: most people regularly pay $10-20 per month for Netflix, Hulu, etc. so those companies can create endless content and make money. All we are asking for is $10-20 a year to help us cover our costs, and keep this work going so we can share it with everyone. Of course, the main point of the cameras is wolf research (and now, a little cougar research on the side) but cool footage of wildlife is a byproduct of having so many cameras out. So help us out and donate here:

Voyageurs Wolf Project

44,123 views • 2 months ago

These look like magic but the "trick" is shown at the end of the video. You can try these animated splats directly on the web on your #Quest3 or #VisionPro by going to I have experienced various types of motion captures or holograms in #Virtualreality and these do look very promising. You also have to consider that to capture something like this, you still need multiple cameras as you can see from the 2 tripods that pop up in the last capture. Do you think that thanks to #AI it will ever be possible to obtain something similar using only one camera 🤔?

These look like magic but the "trick" is shown at the end of the video. You can try these animated splats directly on the web on your #Quest3 or #VisionPro by going to I have experienced various types of motion captures or holograms in #Virtualreality and these do look very promising. You also have to consider that to capture something like this, you still need multiple cameras as you can see from the 2 tripods that pop up in the last capture. Do you think that thanks to #AI it will ever be possible to obtain something similar using only one camera 🤔?

Gabriele Romagnoli

12,229 views • 2 years ago

A 10-minute walk. That's all it takes to transform a real-world neighborhood into a machine-readable digital twin. Captured with an Insta360 X5 and reconstructed by Over the Reality, this environment is no longer just a place on a map. It's spatial data that can power Visual AI, Robotics localization, VPS, XR experiences, and next-generation world models. Today, the OVER community has already created more than 270,000 machine-readable locations like this, forming one of the world's largest 3D datasets for Physical AI, an increasingly valuable asset as demand for high quality real-world data continues to grow.

A 10-minute walk. That's all it takes to transform a real-world neighborhood into a machine-readable digital twin. Captured with an Insta360 X5 and reconstructed by Over the Reality, this environment is no longer just a place on a map. It's spatial data that can power Visual AI, Robotics localization, VPS, XR experiences, and next-generation world models. Today, the OVER community has already created more than 270,000 machine-readable locations like this, forming one of the world's largest 3D datasets for Physical AI, an increasingly valuable asset as demand for high quality real-world data continues to grow.

Over the Reality 🌐

16,642 views • 27 days ago

Christopher Nolan convinced IMAX engineers to completely redesign their cameras just so he could film The Odyssey. "I called my friends at IMAX and I said, 'If ever we're going to make a film entirely in IMAX, this is The Odyssey. So, can you find a way to alleviate the limiting factor?'" "The limiting factor being the cameras are extremely loud." "We've never been able to capture dramatic scenes with an IMAX camera." "What they did is they came up with a way of blimping the camera." "Essentially a high-tech box that you put the camera into and it silences it." "It let us, for the first time ever, get the quiet scenes... and therefore be able to do the whole film."

Christopher Nolan convinced IMAX engineers to completely redesign their cameras just so he could film The Odyssey. "I called my friends at IMAX and I said, 'If ever we're going to make a film entirely in IMAX, this is The Odyssey. So, can you find a way to alleviate the limiting factor?'" "The limiting factor being the cameras are extremely loud." "We've never been able to capture dramatic scenes with an IMAX camera." "What they did is they came up with a way of blimping the camera." "Essentially a high-tech box that you put the camera into and it silences it." "It let us, for the first time ever, get the quiet scenes... and therefore be able to do the whole film."

Nichole

805,940 views • 21 days ago

GS^3: Efficient Relighting with Triple Gaussian Splatting Abstract: We present a spatial and angular Gaussian based representation and a triple splatting process, for real-time, high-quality novel lighting-and-view synthesis from multi-view point-lit input images. To describe complex ap pearance, we employ a Lambertian plus a mixture of angular Gaussians as an effective reflectance function for each spatial Gaussian. To generate self-shadow, we splat all spatial Gaussians towards the light source to obtain shadow values, which are further refined by a small multi-layer perceptron. To compensate for other effects like global illumination, another network is trained to compute and add a per-spatial-Gaussian RGB tuple. The effectiveness of our representation is demonstrated on 30 samples with a wide variation in geometry (from solid to fluffy) and appearance (from translucent to anisotropic), as well as using different forms of input data, including rendered images of synthetic/reconstructed objects, photographs captured with a handheld camera and a flash, or from a professional lightstage. We achieve a training time of 40-70 minutes and a rendering speed of 90 fps on a single commodity GPU. Our results compare favorably with state-of-the-art techniques in terms of quality/performance.

GS^3: Efficient Relighting with Triple Gaussian Splatting Abstract: We present a spatial and angular Gaussian based representation and a triple splatting process, for real-time, high-quality novel lighting-and-view synthesis from multi-view point-lit input images. To describe complex ap pearance, we employ a Lambertian plus a mixture of angular Gaussians as an effective reflectance function for each spatial Gaussian. To generate self-shadow, we splat all spatial Gaussians towards the light source to obtain shadow values, which are further refined by a small multi-layer perceptron. To compensate for other effects like global illumination, another network is trained to compute and add a per-spatial-Gaussian RGB tuple. The effectiveness of our representation is demonstrated on 30 samples with a wide variation in geometry (from solid to fluffy) and appearance (from translucent to anisotropic), as well as using different forms of input data, including rendered images of synthetic/reconstructed objects, photographs captured with a handheld camera and a flash, or from a professional lightstage. We achieve a training time of 40-70 minutes and a rendering speed of 90 fps on a single commodity GPU. Our results compare favorably with state-of-the-art techniques in terms of quality/performance.

MrNeRF

17,786 views • 1 year ago

Awesome footage of a wolf pack erupting in a howl on top of a frozen beaver lodge this past winter! You can hear the howls of the adults and the yipping of the pups, which sound kind of like coyotes. Literally, the only way we keep our 350-400 cameras going year round, capturing footage like this, is via donations. So, if you enjoy our trail camera footage, please donate to our annual fundraiser at the link below—we use these funds to purchase more trail cameras, buy batteries, SD cards, locks, mounts, etc. The trail cameras not only capture cool footage we can share but also provide us with invaluable data to study and understand pup survival, pack dynamics, and many other aspects of wolf ecology. You can support our work here:

Awesome footage of a wolf pack erupting in a howl on top of a frozen beaver lodge this past winter! You can hear the howls of the adults and the yipping of the pups, which sound kind of like coyotes. Literally, the only way we keep our 350-400 cameras going year round, capturing footage like this, is via donations. So, if you enjoy our trail camera footage, please donate to our annual fundraiser at the link below—we use these funds to purchase more trail cameras, buy batteries, SD cards, locks, mounts, etc. The trail cameras not only capture cool footage we can share but also provide us with invaluable data to study and understand pup survival, pack dynamics, and many other aspects of wolf ecology. You can support our work here:

Voyageurs Wolf Project

42,976 views • 2 months ago

New 8K immersive 3D 180 camera is here, designed with usability and affordability in mind. But how does it really compare to cameras like the Canon R5C, CALF Gen 2, and Canon R7: In this review, we take a closer look at SLAM XCAM 3D VR180 camera real-time HDR playback, wide 190° field of view, and other features that could make it an accessible choice for immersive video. Comment “VR180,” and I’ll DM you a FREE side-by-side comparison video to test on your Meta Quest 3 or Apple Vision Pro. #VR180 #immersivetech #3d180

New 8K immersive 3D 180 camera is here, designed with usability and affordability in mind. But how does it really compare to cameras like the Canon R5C, CALF Gen 2, and Canon R7: In this review, we take a closer look at SLAM XCAM 3D VR180 camera real-time HDR playback, wide 190° field of view, and other features that could make it an accessible choice for immersive video. Comment “VR180,” and I’ll DM you a FREE side-by-side comparison video to test on your Meta Quest 3 or Apple Vision Pro. #VR180 #immersivetech #3d180

Hugh Hou

16,704 views • 1 year ago

Scale alone is not enough for AI data. Quality and complexity are equally critical. Excited to support all of these for LLM developers with Snorkel AI Data-as-a-Service, and to share our new leaderboard! — Our decade-plus of research and work in AI data has a simple point: scale alone is not enough. AI success is all about the quality, complexity, and distribution of data—in addition to volume. We’re excited to be powering leading LLM developers with Snorkel AI Expert Data-as-a-Service, our white glove service for custom, expert-level AI datasets—and to now preview some of what we’re building via our new Expert Data Leaderboard (🔗 in 🧵) + upcoming OSS dataset releases! Snorkel Expert Data-as-a-Service is built to meet the rapidly evolving data needs of the agentic AI world—where success is built on the quality, complexity, and distribution of datasets, in addition to size and scale. This kind of high-quality, frontier AI data can only come from a union of technology and human expertise. With Snorkel Expert Data-as-a-Service, we’re powering frontier LLM developers across agentic, expert knowledge, reasoning, coding, multi-modal, and other task types via the combination of these two key components: - (1) The Snorkel Expert Network: A global team of subject matter experts focused wholly on specialized knowledge–spanning thousands of topics in STEM/academic, vertical/professional, and consumer/lifestyle domains. - (2) Snorkel AI Data Development Platform: Our unique programmatic data curation and quality control platform, accelerating and improving expert authoring and review through principled techniques developed over the last decade of R&D. Now: we’re incredibly excited to showcase some of the power of Snorkel Expert Data-as-a-Service via the new Snorkel Leaderboard—putting frontier models to the test in complex, agentic, and reasoning settings inspired by real industry scenarios (not esoteric puzzles)! We’ll be releasing new leaderboards and accompanying expert-verified open source datasets (coming soon!) regularly. To start, we’re sharing three initial ones in preview: - SnorkelFinance: Q&A over financial documents requiring agentic tool-calling and reasoning - SnorkelUnderwrite: Agentic insurance tasks requiring industry-specific reasoning and tool use - SnorkelSequences: Mathematical tasks requiring compositional multi-step reasoning

Scale alone is not enough for AI data. Quality and complexity are equally critical. Excited to support all of these for LLM developers with Snorkel AI Data-as-a-Service, and to share our new leaderboard! — Our decade-plus of research and work in AI data has a simple point: scale alone is not enough. AI success is all about the quality, complexity, and distribution of data—in addition to volume. We’re excited to be powering leading LLM developers with Snorkel AI Expert Data-as-a-Service, our white glove service for custom, expert-level AI datasets—and to now preview some of what we’re building via our new Expert Data Leaderboard (🔗 in 🧵) + upcoming OSS dataset releases! Snorkel Expert Data-as-a-Service is built to meet the rapidly evolving data needs of the agentic AI world—where success is built on the quality, complexity, and distribution of datasets, in addition to size and scale. This kind of high-quality, frontier AI data can only come from a union of technology and human expertise. With Snorkel Expert Data-as-a-Service, we’re powering frontier LLM developers across agentic, expert knowledge, reasoning, coding, multi-modal, and other task types via the combination of these two key components: - (1) The Snorkel Expert Network: A global team of subject matter experts focused wholly on specialized knowledge–spanning thousands of topics in STEM/academic, vertical/professional, and consumer/lifestyle domains. - (2) Snorkel AI Data Development Platform: Our unique programmatic data curation and quality control platform, accelerating and improving expert authoring and review through principled techniques developed over the last decade of R&D. Now: we’re incredibly excited to showcase some of the power of Snorkel Expert Data-as-a-Service via the new Snorkel Leaderboard—putting frontier models to the test in complex, agentic, and reasoning settings inspired by real industry scenarios (not esoteric puzzles)! We’ll be releasing new leaderboards and accompanying expert-verified open source datasets (coming soon!) regularly. To start, we’re sharing three initial ones in preview: - SnorkelFinance: Q&A over financial documents requiring agentic tool-calling and reasoning - SnorkelUnderwrite: Agentic insurance tasks requiring industry-specific reasoning and tool use - SnorkelSequences: Mathematical tasks requiring compositional multi-step reasoning

Alex Ratner

495,851 views • 1 year ago

The era of manually analyzing data will come to an end. AI can now do a lot of this automatically. It's a huge time saver. I do data analysis for a living, and I'm a huge fan of writing Jupyter notebooks to do it all, but it's now hard to justify manually writing code that you can generate in a few seconds. I still check everything manually, but breaking down datasets into tables and charts is now 10x easier than it's ever been. Here is a video where I'm using Retool. I load a dataset and generate a few charts as quickly as I can think. The speed at which we can go from one idea to a working solution is astonishing.

The era of manually analyzing data will come to an end. AI can now do a lot of this automatically. It's a huge time saver. I do data analysis for a living, and I'm a huge fan of writing Jupyter notebooks to do it all, but it's now hard to justify manually writing code that you can generate in a few seconds. I still check everything manually, but breaking down datasets into tables and charts is now 10x easier than it's ever been. Here is a video where I'm using Retool. I load a dataset and generate a few charts as quickly as I can think. The speed at which we can go from one idea to a working solution is astonishing.

Santiago

70,812 views • 10 months ago

More awesome footage of one of the most ethereal forest creatures: the lynx. We captured more daytime footage of lynx this past winter than in any previous winter, in large part, because we had many more cameras out this winter compared to previous years (had ~350 cameras out this winter). And more cameras means, we get more neat observations like these (plus higher-quality data on wolves in the area, which is the main reason we deploy so many cameras). Note: this is the 2nd video of our “best" lynx footage from this winter. See our post on Monday for the 1st video if you haven’t watched that one yet. Both have some pretty neat sequences in our opinion! Huge thanks to the >8,700 individual donors whose generosity has allowed us to deploy a large number of trail cameras and capture neat footage like this to share with everyone.

More awesome footage of one of the most ethereal forest creatures: the lynx. We captured more daytime footage of lynx this past winter than in any previous winter, in large part, because we had many more cameras out this winter compared to previous years (had ~350 cameras out this winter). And more cameras means, we get more neat observations like these (plus higher-quality data on wolves in the area, which is the main reason we deploy so many cameras). Note: this is the 2nd video of our “best" lynx footage from this winter. See our post on Monday for the 1st video if you haven’t watched that one yet. Both have some pretty neat sequences in our opinion! Huge thanks to the >8,700 individual donors whose generosity has allowed us to deploy a large number of trail cameras and capture neat footage like this to share with everyone.

Voyageurs Wolf Project

108,930 views • 1 year ago

Today, Search Live is available in English in the U.S. — no Labs opt-in required. ✨ You can now have a free-flowing, back-and-forth conversation with Search using your voice and phone camera. Live in Search combines the power of Google Lens and AI Mode to see and accurately identify objects in front of you. This allows Search to understand exactly what you are looking at and help you find the most relevant information across the web to respond to your questions in real time. Try it today by tapping Live in the Google app (iOS and Android) or Lens, and let us know what you think!

Today, Search Live is available in English in the U.S. — no Labs opt-in required. ✨ You can now have a free-flowing, back-and-forth conversation with Search using your voice and phone camera. Live in Search combines the power of Google Lens and AI Mode to see and accurately identify objects in front of you. This allows Search to understand exactly what you are looking at and help you find the most relevant information across the web to respond to your questions in real time. Try it today by tapping Live in the Google app (iOS and Android) or Lens, and let us know what you think!

Google AI

48,572 views • 10 months ago

If I learned anything over 5 years as a reporter, it’s this: the best research is firsthand. That’s why we’re doing a review of the latest Solana phone for Bitwise There's a lot of skepticism about Seeker. Some doubt it will work, or that the crypto elements will be helpful. They say it'll be a nice paperweight. Why are we doing this? There's two ways to research things in crypto. You can dig into the onchain data and talk to the experts. Or you can just do it. You can trade on hyperliquid. You can loop for yield. You can travel to Argentina and try to use stablecoins to pay for things. We do a lot of the traditional research at Bitwise, but sometimes, you gotta get your hands dirty. That's what I'm doing with this phone.

If I learned anything over 5 years as a reporter, it’s this: the best research is firsthand. That’s why we’re doing a review of the latest Solana phone for Bitwise There's a lot of skepticism about Seeker. Some doubt it will work, or that the crypto elements will be helpful. They say it'll be a nice paperweight. Why are we doing this? There's two ways to research things in crypto. You can dig into the onchain data and talk to the experts. Or you can just do it. You can trade on hyperliquid. You can loop for yield. You can travel to Argentina and try to use stablecoins to pay for things. We do a lot of the traditional research at Bitwise, but sometimes, you gotta get your hands dirty. That's what I'm doing with this phone.

Danny Nelson

61,480 views • 10 months ago

For me, this was the moment when Halo went from being a War Story with futuristic elements to a Sci-fi Superhero Story. It’s an impossible shot. The camera flies around and through objects, time speeds up and slows down at random, the camera anticipates events that are scripted instead of reacting to capture them, there is no sense of space or proximity or cause and effect... It's like they gave the camera to Superman. That's all fine in a Superhero Story, but it breaks the stakes of a War Story. If the audience perspective is not grounded, then the stakes of the story are not grounded. It's exciting and flashy, but has as much impact as a kid playing with dolls.

For me, this was the moment when Halo went from being a War Story with futuristic elements to a Sci-fi Superhero Story. It’s an impossible shot. The camera flies around and through objects, time speeds up and slows down at random, the camera anticipates events that are scripted instead of reacting to capture them, there is no sense of space or proximity or cause and effect... It's like they gave the camera to Superman. That's all fine in a Superhero Story, but it breaks the stakes of a War Story. If the audience perspective is not grounded, then the stakes of the story are not grounded. It's exciting and flashy, but has as much impact as a kid playing with dolls.

Aozolai

317,793 views • 9 months ago