First time I saw this idea was Berg's "Here and There" poster from 2009 (a year before Inception) though they acknowledge a bunch of earlier inspirations...
A computer generated 3-dimensional map display includes a reference icon identifying an observation point in the geographic area defined by the map, the map being presented on a curved surface which extends from a first plane to a second plane which is substantially orthogonal to the first plane. The observation point is located in the first plane. The display is configured such that the height of landmarks and buildings in a defined foreground of the reference icon are presented normal to the first plane, the height of land marks and buildings at a defined far distance are presented normal to the second plane and the height of land marks and buildings in a defined middle ground are presented normal to the non-linear portion of the curved surface. The geographic area represented can be translated along the curved surface to provide a perception that the reference icon is roaming through the displayed geographic area.
By no other than "Here and There" Schulze and Webb Ltd later rebranded to BERG.
The generic use case of a multiperspective camera method for that was arguable covered in a 1997 Disney paper latest.
The patent application is not yet publically available, so I can't tell what aspects of the system they are trying to protect. They probably have some narrowly tailored features that they can protect. Maybe related to the rendering, or some toggle button feature of the display. But, as to the broad concept of a curved hybrid combination of street view and overhead view, I agree that they will not be able to protect that aspect.
Especially that it looks like to be an European patent and it is not clear what wouldn't be "software" in their patent.
But if it's not vibecoded landing page, they might also know that they have a dummy patent request that will probably be rejected but that probably in so many month that they can use the patent pending to get commercial and investment traction till then.
Yep. First time I saw this idea was Animal Crossing: Wild Word's "rolling log" effect in 2005, albeit in reverse. (So you get a top-down view and a horizon view at the same time). Second time I saw this specific idea was a Minecraft vertex shader in 2013 with the "Acid Shaders" vertex transform shaders.
I remember working on something like this as early as 1999! It was for a small European company, sadly we weren't able to put it out due to licensing restrictions at the time.
This is a really good proof of concept. I would use this.
Something about the moment of the turn itself isn’t working for me though. Just before the turn you have essentially no information about the route ahead, until after the turn is completed. This would make a sequence of consecutive turns difficult to navigate.
This is exactly why on traditional gps maps I prefer it to be very very zoomed out. It's far too common that mapping tools make something like a sudden second turn invisible to you until it is too late to switch lanes.
I find that Apple Maps is typically pretty smart with its automatic zoom. For a typical intersection with just one turn to navigate, it will zoom in to give you a good sense of the lanes, stop signs/lights, etc. But when there are consecutive turns to navigate, or a heavily curved section of road (like highway on-ramps) it will zoom out and even rotate the map to show that upcoming context.
It might be fine for pure navigation to addresses but business addresses and other metadata are often missing or out of date because people only bother to update Google Maps. Also that’s where all the reviews are.
I don’t want to use two different apps for navigation vs finding businesses.
I've used it almost exclusively since around 2019, but that has been in the Bay Area and LA area where the data has presumably been very good relative to other areas.
I remember for at least a while people would complain about how GPS navigation ruined their sense of the local area. Probably stopped hearing about it because people have just surrendered to the apps, but I think part of the reason is also that by default they were just zoomed in way too close. If you forced them back out a ways, I was actually learning a lot about my local area. Like, there's all these subdivisions around, you know, and from the outside who knows what connects to what, I'm not going to just duck in to a subdivision and physically navigate the streets to see how they're all connected, but driving around with the navigation made it much more clear... when it was zoomed out somewhat.
I had similar thoughts, but this is a largely solved problem: even headlights in my Volvo turn as I turn the wheel, and it should be easy to change rendering direction to an "average" of road tangents for the next say 50m (would depend on the speed and would need experimenting to know the best value).
This is also a problem with Google Maps navigation. It is always showing the perspective looking straight ahead of your current direction, even when you're approaching a sharp turn. Instead, the perspective should be along a vector connecting your current location with the point where the navigation's path crosses the edge of the display, pointing straight up.
Yes, this is how it's currently implemented. My proposal is "up is the intersection of the path with the edge of the screen". Here's a mock comparison of the current behavior with my proposal: https://i.imgur.com/AmUnzfS.jpeg (The mocked perspective is wrong because I just cut out and rotated the image, but hopefully you get the idea)
So why not use 2D maps, have people lost the ability to read North up maps? I swear ever since paper maps have not been used in cars, people have gotten spatially dumber.
2D map would work the same way with my proposal. The "north is up" part is the issue for me, I'd like to have the rotation oriented in such a way that (the visible part of the path to) my destination is up.
I love north-up maps because I find they give me way better spatial context. My truck’s nav system gives north-up as an option but my phone and my truck map both suffer from the same problem on different axes. My phone sits in its holder vertically, so I have way more north-south context than east-west. The truck screen is landscape, same problem north-south! A square map north-up would be amazing!
This looks pretty neat but also distracting and not very convenient. A useful follow mode should provide a view into the next several dozen seconds of road conditions. This projection forces road sections after sharp turns to go off screen, and does nothing to compensate for that (e.g. by rotating the view), so the distance and time of useful prediction changes constantly. Since you're already willing to give the driver nausea, consider unwrapping 90 degree turns into something more straight until they're close enough.
I think it'd be pretty easy to fix, too. Have a point on the line that represents "10 seconds from now" and orient your camera along the line between that and the point representation of your car, with slow, smooth panning so as not to get jerky transitions. If "10 seconds from now" is within a certain range of the car, it ought to just default to whatever its doing now.
I feel like video game logic would be useful here where a view like this would replace the main view and the “macro view” would be a minimap in the corner.
I think where the actual distraction is comes from the level of detail. It would actually benefit from just having silhouettes or lower resolution “suggestions” of what’s there. It’s maybe just the buildings on the corners of turns or specific landmarks you’d want fully rendered so people could follow ambiguous directions like “Turn left at church, but not the Methodist church the Episcopal one!”
Yea, I felt like I was perpetually at the top of a roller coaster and going over the hump, but never finding the bottom. And I like roller coasters. But no thanks. Not even sure what the benefits of this view really are.
Nice demo, but can we talk about how this is patented. Should it be? Should something like this actually get granted.
While it's neat, it's neat on the order of a UX idea someone might just come up with at standup one day. Like, cool, worth an A/B test, but worth a patent? I don't think so.
Great idea and demo. My main gripe when using Google Maps in the car is lack of information about the next turns — when do I change lanes, and to which one? This could be natively addressed here, especially if the blue navigation line was a little thinner and showed which lane(s) the driver needs to take.
> Google Maps (at least in the US, in my experience) is really great about telling you what lane to be in.
Having recently driven in Boston[1], the problem is that Google Maps tells you what lane you need to be in when the route divergence happens... it doesn't tell you what lane you need to be in to get to that lane.
So I would see guidance like be in the first or second lane (from the left) to make this left turn and all seemed well since I was in the second lane from the left. Then two more lanes merged in from the left and I had to get over two lanes and ended up on a two mile onramp for the highway to New Hampshire.
I see the same things locally, but it's not as obviously bad because I'm familiar with the local highways and also most of the local highways aren't as space constrained so there's almost always plenty of time to merge.
[1] I don't recommend it... we should have driven to the airport and returned our car and taken transit back to our hotel
Where in the world are these road owners, and do they really eve provide this kind of information to mapping companies?
Here in the States, most of the roads are "owned" by the people and our various thousands of governmental jurisdictions do not tell anyone about these kinds of details in any uniform way.
That's also where this map would help, it can definitely degrade gracefully. If I know I have two right-hand turns coming, I can take the decision for myself.
This information is all derived from overhead (satellite and aerial) imagery combined with street level imagery (like from Google street view). No road owners required.
I find these comments fascinating. Do people just switch off and become completely incapable of using the road when using GPS? You can use navigation and your ability to use roads, read road signs etc at the same time!
This is cool. May I suggest that you make the standard deformation much less spherical? I do not get nauseous that easily and it made me a little nauseous, feeling like I was falling. I would not want to watch that while driving in a car. But 1/3 of the way from ‘flat’ to that would be super useful.
I'm with you on this. I think this could actually be really useful but the effect is way too extreme. Additionally, the animation is quite stuttery on turns --- might just be my machine though.
Same. I don't get motion sick, and this made me feel uneasy. It would make my wife motion sick instantly. It's an interesting concept though, and I can see the utility of it.
Interesting. I had a similar idea around a decade ago.
My thoughts from the time: if it's just [something like the linked implementation], there's a limited view of the way ahead, so it's better if you can give it a non-linear scale.
Upcoming turns also hide a lot of the real way ahead, so it's better to if you can increase horizontal visibility too.
How do you have non-constant scale on both axes? How to make this useful for navigation?
I may play with this idea again given that vibe coding has given me time for such experiments.
I don't really see how this could ever be better than what Google maps already does. It's neat and cool and stuff, it's just not useful. And you need a way beefier computer in your car to run it.
It may well be worse. This is one reason I didn't judge doing experiments to find out to be a good use of my time before vibe coding became a thing.
However, I also wouldn't bother with the full 3D rendering, I'd just use a normal map. There's several reasons why normal maps don't tend to include every bush, every letter box on every building.
That's nice however I'll be surprise you can patent a few transformation matrices for an effect present in the public Zeitgeist since more than a decade. The core of the technique being Gaussian Splatting I guess I wonder how you'll get the data, except maybe SFM-ing your way through a worldwide collaborative effort.
I suppose Google could do something like that with their own data.
I don't really know why, but the thought of my (elderly) mother using this made me laugh almost as hard as the El Risitas guy[1]. I adore it, and it also should never ever be used in any car GPS scenario. It's far too distracting, and puts far too much emphasis on the places you are not going.
I could see it being useful for a more exploratory style of map, however. Games like Pokemon Go come to mind as a place where this would be pretty cool.
> far too much emphasis on the places you are not going
I have this issue with literally every modern nav system. They all make their fundamental display a line marked on a map, and most of the information on that map is irrelevant.
More than 20 years ago I had some kind of Verizon flip-open phone with a QWERTY keyboard. No idea what model, but it had this fabulous nav interface that just showed big arrows showing your next two turns and a schematic diagram of what the next turn would look like. I wish I could get something like that, but with Google's data as the back end.
I tried three times on my 5-year-old phone. First time got stuck at 90% loading. Second time the tab crashed at 90% loading. Third time Chrome itself crashed once loading finished.
I miss such a landmark feature in all available routing apps and think it will become standard in selfdriving cars where there is more possibility for the driver to look around the landscape
The concept is good, the implementation not so much, at least not on desktop. Mouselook doesn't make sense here; A and D should turn, not sidestep, which you can't do in a car. You do show a car right at the outset so people are going to imprint on that. Lack of arrow key support also alienates non-gamers.
This is surprisingly useful. As I was imagining myself behind thr wheel I felt it was more clear as to when/where the next turn is. Probably because it combines top down flat with 3d spatial vs just one or the other. This was really good. I want it.
Absolutely honest question: what is the advantage of this over a top view navigation with the orientation following the orientation of movement? Am I missing something?
If you've ever been confused about which turn is which with top down navigation, then this is for you. While driving a car, being able to at a glance know exactly what turn is what would have saved me countless wrong turns, my parents probably doubly so.
One classic: there's a fork in the road on the way to my parents. One side is "straight", the other side is an exit. I took the wrong, exit side of that several times while my app said to go straight before I just knew the route and didn't need GPS at all. Top-down isn't as good in these cases because of densely packed roads and back to back exits.
That said, it's "good enough" and I doubt this takes off.
while I think that the widgets telling you which lane to put yourself in largely solve this kind of intersection confusion, I understand the problem you are describing
but this seems to me just a topdown and "third person" at the same time through the distortion, I don't see how it helps with knowing exactly where to turn, is it because it shows the way further down the turn (like, sideways after the turn in the demo) ?
For me, confusion in top-down navigation happens either because there's no bitmapped imagery being displayed, forcing me to guess where the turn is by mentally wrapping vague 2D geometry over the scene viewed through my windshield, or because the trace isn't drawn precisely enough over the imagery that is available.
I don't see much added value with this particular nausea-inducing approach. It falls into the category of "You should've thought less about whether you could and more about whether you should."
You can calibrate navigation to visual landmarks, i.e. you can see the turn coming up and match it to what you see out your window. But you also still get an overhead "what lies ahead" map for planning your route further out. This is an interesting concept, maybe a hybrid approach would be useful, a simple overhead streets map as default that swaps to the 3rd person view when you approach your turn.
Maybe nothing here is patentable. We don't know yet.
All we know is that they've applied for a patent. If the patent they seek is granted, then it will be published and there will be no mystery about what it covers. :)
Isn't it just static files? Use a cdn like cloudflare or AWS CloudFront or depending on size - if you can break it into chunks of like 50mb each, GitHub pages (free)
200 people or 100k people wouldn't matter then, it'll be cached.
It will also make the download based almost entirely on client download speed.
But estimating distances is hard, as is estimating road size. You can only plan one straight line ahead, and there's no "glance how many blocks before I make a left turn". Now it's "scan and process the entire straight terrain section with eyes recognizing crossings to get the count" Alternatively, I'll have to task switch all the time to it to not miss on the turn. Some hybrid map visual overlay that makes reading it at a glance easier would go a long way here.
Also any points shown (e.g. destination, POIs) that are more than a few degrees away from straight ahead appear to be next to you, but following you which is quite confusing.
Not sure if this is meant to be practical, but I think this would work better if the upcoming turn was centered in the view once it’s within a certain range, as a kind of anticipation. This helps prepare for the jarring shift in perspective and, in case there’s multiple turns in short succession (as there might be when exiting a highway, for instance), gives you a visual heads-up about what’s next.
Beautiful perspective if you need to take a left in 500 meters. Not so great if you have to take a left in 3 kilometers because that would just not be visible.
Now of course you could change the parameters so that the upper screen border is 3 kilometers. But normal standard perspective is even better, because it can show you up to infinity. The inherent concept here is that there is a cut-off at the top, and that there is always a lot of unnecessary stuff near you whizzing by to look at / dopamine kicks. So you would have to constantly look at the screen - instead of at the road.
Navigon (among others) back in 2009 showed a perspective that balanced between showing the next junction and the next few kilometers. And leaving out anything unnecessary. I like that better. Example: https://www.notebookcheck.com/fileadmin/_processed_/csm_Garm...
The trick would be adapting relatively, which a lot of apps already do... zoom/flatten out if it's a long stretch of nothing, back in when you start approaching more complicated sections.
It might be useful to be able to configure the map viewing angle at the top and bottom of the frame, perhaps even the degree of curvature between them.
I say this because I wonder if a shallower gradient (slightly more top-down at the bottom and slightly more angled at the top) would be more understandable and familiar while providing almost as much information. Not as cool-looking, but potentially more useful as a real navigation aid.
thanks. Im still tweaking the map and camera settings, its far from perfect yet. Maybe next demo update I'll include some of the sliders to play around with
I think this takes the record for top thing that I expected to be wildly unintuitive and turns it to be incredibly intuitive. I absolutely love this and want it for real. It gives me a road view and an aerial view and transitions smoothly. Needs polish and to somehow be accessible in seconds for everywhere but I would default settings this in my car if it was there.
A problem of open GeoData is very much their coverage and availability in the EU. While the countries and their survey offices are required to release high value datasets in standardized form, some simply don’t. In Germany we have 16 states and every one of them has their own standard and pipelines for generating that data, and everyone releases it differently. Take LiDAR: Bavaria should be distributing it through OpenData portals and is required to by the EU, but they simply ignore that and charge 8000€ for it. Building international or even EU wide geo-projects from OpenData like https://hillsha.de is difficult because of that, you never get full coverage.
But as a tool for mapbuilders for small areas or planned tours i think this could be great
Very cool, after watching a minute it grew on me and I can see why the perspective is useful.
My immediate thought is that this is what is would be like to drive in The Spindle in Neuromancer or any similar space environment that’s spinning for artificial gravity.
As a side effect, this is a neat way to visualize Gaussian splatting! As you go further away from the main navigation route, the quality of the rendering continuously degrades, until it eventually becomes sparse ellipsoids (i.e., you can see the individual splats).
A variant of this but taking place on an O'Neill Cylinder would be amazing. I've always been captivated by the idea of the horizon curving up in front of you.
I almost immediately click-dragged into a nearby building by accident, and am fascinated by how this engine attempts to extrapolate the inside of the building using only information about the outside. It's kind of surreal, and slightly more convincing than it feels like it should be at times
In the demo, the "Destination" and "Point of Interest" seem to be floating around randomly. I'd expect them to point out the things they're labeling, but their movement seems to bear little correlation to anything.
I also experienced vertigo, but I'm particularly prone to vertigo induced by 3D scenes. Still, I think this is the first time a screen as small as my phone did it.
Nice work! This sort of reminds me of a game I saw demo'd on HN about a year ago, where you could walk and talk to the people and do some tasks- the difference being - in that game the world felt like a convex sphere and here it is concave.
Very cool. This would actually save me from those terrible times when the navigation tells me to turn, but not which lane is best for the next turn after that. Now I could visualize it clearly.
That being said, it is a little dizzying to watch.
At first I thought this was a cool gimmick, but this seems like it would be really helpful. You get to see the angle of the curve from the driver's perspective and the overhead view at the same time!
That was surprisingly nice. I get to see the far distance on the map. Something has to be done about turns, though. They’re far too whirly for me. Good demo!
I don't remember a map from the movie so I didn't know what to expect. Instantly I love this. I actually wish I could have this rather than street view in Google Maps.
This is really cool. I'm also reminded of a game on SGIs called Mekton. Mekton was originally a mecha-combat RPG designed by Mike Pondsmith, the creator of Cyberpunk (2040, etc.); it was made into a multiplayer video game using OpenGL that was set on a ring-shaped space colony. When you land, you can see the landscape curve upwards ahead of you, as you're on the inside of the ring.
Unusual. Nice concept. I ran the demo. How does it select the vehicle used in the demo? Is that something that I can change? Overall I ran it on my old workstation desktop under LibreWolf with no issues.
this looks pretty cool.
since it curves the the 3d map, what would it look like when you are driving through downtown highrises? would the tops of the buildings clash into each other?
Even crazier idea would be the possibility of 'normal bullets' that travel along the curve, and 'special bullets' that travel on a direct diagonal via warped space & time.
This is pretty intuitive and useful, I think. I would suggest swapping to an angle equal to half the angle of the turn once your approach is within a certain distance, to show the user what is coming up around that turn. Frequently you need to turn then turn, so seeing 'ahead' a bit would be cool
Navigation needs to be low cognitive overhead to have mass appeal. There's a reason why people do not navigate with satellite or photo real maps. This is an interesting projection and visualization but doesn't have much practical value.
Hold down secondary mouse button, typically the right one, and move the mouse. (If you don't have a mouse attached, you might not be able to perform this action.)
Adding two existing features together (gps mapping on an inverted sphere) doesn't seem to be the basis of a patent to the rules of incremental innovation.
Does anyone else hate how Google Maps zooms in right before a turn?
Yeah go ahead and completely change the scale of what I’m looking at a night quite predictable time just as I’m about to need the map the most to make a turn on a street I’m not familiar with
Absolutely fucking brilliant idea
This is why I use Apple Maps and thankfully in the Bay Area it’s more than at parity, it’s frankly better
All that said, this UI is actually awesome - just needs a little tweaking for the turn animation but I’d 100% use this - incredibly intuitive
Cool idea, but it feels more suited for pedestrians or cyclists moving at a slower pace. Not sure this format would work well in a car, since you can't really see the whole road ahead.
At 100 km/h, you need to grasp the route and upcoming turns quickly and this view might not give you enough time to get in the right lane.
Still, props to them for trying something new.
https://berglondon.com/blog/2009/05/04/here-there-influences...
reply