#surfelation | Explore Tumblr Posts and Blogs

linkgreys · 2 years

Text

Forza horizon 5 xbox

Forza horizon 5 xbox 1080p#

Forza horizon 5 xbox full#

Forza horizon 5 xbox series#

That leads us onto the other aspect of FH5 on Xbox One that emphasises quality - the 30fps target frame-rate barely budges. In fact, Playground deploys what it calls 'DRS Plus', dropping back fidelity on cube-map reflections and shadow updates before lowering resolution. Dynamic resolution scaling to 810p is there (DRS being a first for Forza titles as far as I'm aware) but it's a tool of last resort as opposed to a core element of the visual make-up.

Forza horizon 5 xbox 1080p#

It also helps that the developer has retained 1080p resolution and 4x MSAA anti-aliasing: Forza titles have always looked pristine in this regard and FH5 is no exception. Because it's content complete, because the key features are all included and because the result has been tailored by Playground to mask as many of the limitations as possible, it's still a worthy sequel. In fact, most of my initial play-testing was on the base Xbox One, simply because I wanted to see how Playground did it. It's not so much of an issue in closed-circuit racing but it's in the open world that the limitations of the console start to bite.

Forza horizon 5 xbox full#

Actual in-world detail takes a significant hit with level of detail transitions more evident - barrelling over the open world at high speed, the pop-in is obvious and sometimes it seems that the 2D 'imposters' for world detail never get to transition to full 3D, and if they do, shading seems minimal. The ground has far fewer layers to it compared to the other consoles, looking rather flat at slow speeds (it's far less noticeable in the thick of gameplay, however). Closer to the player, that's where the cuts are more evident. Far-off detail is still rendered to an impressive degree, and as it's a persistent aspect of the scene's visual make-up, it has to be - it sells the scale of the open world.

Forza horizon 5 xbox series#

Probably the most noticeable difference at the core system level is foliage lighting: Playground developed a new system to simulate how light passes through vegetation and it's present on all systems, except Xbox One.Ī key aspect of Forza Horizon 5's visual appeal is its sheer density and its massive vistas, something Xbox One cannot hope to deliver to the same level of fidelity as its Series counterparts. The brand new surfel-based global illumination system - that's there too. It's of a much higher resolution on Series X, but it's still there on Xbox One - with an added post-process pass to eliminate low res artefacts. Proper volumetric lighting is added, for example. The core technologies developed for Forza Horizon 5 look phenomenal on Series X but almost all of them are present in a lower precision form on the base Xbox One. Essentially and perhaps not surprisingly, it's all about scalability. The set-up I saw is testament to the time and effort Playground has put into ensuring a decent experience for every console. Watch on YouTube Everything you need to know about how Forza Horizon 5 scales across every Xbox One and Xbox Series console. Xbox One X? I didn't see that but I needn't have worried, it's a fitting send-off for the Microsoft's first 4K console. From there, moving from left to right, I could see how each console delivered the scene: Xbox Series X quality and performance, Series S equivalents, then finally, the base Xbox One. Using one controller and a network of Xbox consoles, I was able to see the demanding jungle stage of the intro drive playing out in real-time across five of the six iterations of the game, with the debug camera used to zero in on the various rendering techniques Playground had developed for the game. Quite how Playground would deliver this was always the crucial question and it was very satisfying to visit the studio a few weeks back to see exactly how it was done, my tour kicking off with a look at a remarkable cross-platform comparison system the studio developed. Admittedly, we were sceptical about its chances but Playground Games was always optimistic - and having put every version through its paces, the studio has delivered. However, this is a cross-gen release: somehow, Forza Horizon 5 has to run on last-gen machines and still live up to the expectations of quality expected from a first-party studio production. We called it an Xbox Series X masterpiece and that isn't hyperbole - Playground Games has delivered a phenomenal game and a sensational audio-visual experience for its flagship console.

#Forza horizon 5 xbox

0 notes

wsmith215 · 4 years

Text

Waymo is using AI to simulate autonomous vehicle camera data

Waymo says it’s beginning to leverage AI to generate camera images for simulation by using sensor data collected by its self-driving vehicles. A recent paper coauthored by company researchers including principal scientist Dragomir Anguelov describes the technique, SurfelGAN, which uses texture-mapped surface elements to reconstruct scenes and camera viewpoints for positions and orientations.

Autonomous vehicle companies like Waymo use simulation environments to train, test, and validate their systems before those systems are deployed to real-world cars. There are countless ways to design simulators, including simulating mid-level object representations, but basic simulators omit cues critical for scene understanding, like pedestrian gestures and blinking lights. As for more complex simulators like Waymo’s CarCraft, they’re computationally demanding, because they attempt to model materials highly accurately to ensure sensors like lidars and radars behave realistically.

In SurfelGAN, Waymo proposes a simpler, data-driven approach for simulating sensor data. Drawing on feeds from real-world lidar sensors and cameras, the AI creates and preserves rich information about the 3D geometry, semantics, and appearance of all objects within the scene. Given the reconstruction, SurfelGAN renders the simulated scene from various distances and viewing angles.

Above: The first column shows surfel images (more on those below) under novel view, while the second column is the synthesized result from SurfelGAN. The third column is the original view.

Image Credit: Waymo

“We’ve developed a new approach that allows us to generate realistic camera images for simulation directly using sensor data collected by a self-driving vehicle,” a Waymo spokesperson told VentureBeat via email. “In simulation, when a trajectory of a self-driving car and other agents (e.g. other cars, cyclists, and pedestrians) changes, the system generates realistic visual sensor data that helps us model the scene in the updated environment … Parts of the system are in production.”

VB Transform 2020 Online – July 15-17. Join leading AI executives: Register for the free livestream.SurfelGAN

SurfelGAN makes use of what’s called a texture-enhanced surfel map representation, a compact, easy-to-construct scene representation that preserves sensor information while retaining reasonable computational efficiency. Surfels — an abbreviated term for “surface element” — represent objects with discs holding lighting information. Waymo’s approach takes voxels (units of graphic information defining points in 3D space) captured by lidar scans and converts them into surfel discs with colors estimated from camera data, after which the surfels are post-processed to address variations in lighting and pose.

To handle dynamic objects like vehicles, SurfelGAN also employs annotations from the Waymo Open Dataset, Waymo’s open source corpus of self-driving vehicle sensor logs. Data from lidar scans of objects of interest are accumulated so that in simulation, Waymo can generate reconstructions of cars and pedestrians that can be placed in any location, albeit with imperfect geometry and texturing.

One module within SurfelGAN — a generative adversarial network (GAN) — is responsible for converting surfel image renderings into realistic-looking images. Its generator models produce synthetic examples from random noise sampled using a distribution, which along with real examples from a training data set are fed to discriminators, which attempt to distinguish between the two. Both the generators and discriminators improve in their respective abilities until the discriminators are unable to tell the real examples from the synthesized examples with better than the 50% accuracy expected of chance.

The SurfelGAN module trains in an unsupervised fashion, meaning it infers patterns within the corpora without reference to known, labeled, or annotated outcomes. Interestingly, the discriminators’ work informs that of the generator — every time the discriminators correctly identify a synthesized work, they tell the generators how to tweak their output so that they might be more realistic in the future.

Promising results

Waymo conducted a series of experiments to evaluate SurfelGAN’s performance, feeding it 798 training sequences consisting of 20 seconds of camera data (from five cameras) and lidar data along with annotations for vehicles, pedestrians, and cyclists from the Waymo Open Dataset. The SurfelGAN team also created and used a new data set called the Waymo Open Dataset-Novel View — which lacks camera images but starts from scenes and renders surfel images from camera poses perturbed from existing poses — to create one new surfel image rendering for each frame in the original data set. (The perturbations arose from applying random translations and yaw angle.)

Finally, Waymo collected additional sequences — 9,800 in total, 100 frames for each — of unannotated cameras images and built a corpus dubbed Dual-Camera-Post Dataset (DCP) to measure the realism of SurfelGAN-generated images. DCP deals with scenarios where two vehicles observe the same scene at the same time; Waymo used data from the first vehicle to reconstruct scenes and render the surfel images at the exact poses of the second vehicle, producing around 1,000 pairs for judging pixel-wise accuracy.

The coauthors of the paper report that when SurfelGAN-generated images were served to an off-the-shelf vehicle detector, the highest-quality synthesized images achieved a metric on par with real images. SurfelGAN also improved on top of the surfel renderings in DCP, producing images closer to real images at a range of distances. Moreover, the researchers demonstrated that images from SurfelGAN could boost the average precision (i.e., how close estimates from different samples were to each other) of a vehicle detector from 11.9% to 13%.

Waymo notes that SurfelGAN isn’t perfect. For instance, it’s sometimes unable to recover from broken geometry, resulting in unrealistic-looking vehicles. And in the absence of surfel cues, the AI exhibits high variance, especially when it tries to hallucinate patterns uncommon in the dataset, like tall buildings. Despite this, the company’s researchers believe it’s a strong foundation for future dynamic object modeling and video generation simulation systems.

“Simulation is a vital tool in the advancement of self-driving technology that allows us to pick and replay the most interesting and complex scenarios from our over 20 million autonomous miles on public roads,” the spokesperson said. “In such scenarios, the ability to accurately simulate the vehicle sensors [using methods like SurfelGAN] is very important.”

Source link

The post Waymo is using AI to simulate autonomous vehicle camera data appeared first on The Bleak Report.

from WordPress https://bleakreport.com/waymo-is-using-ai-to-simulate-autonomous-vehicle-camera-data/

#Coronavirus News

0 notes