This morning I came across this cool new horror comedy film called “The Final Girls”. It’s a 2015 film that will be released in the United States on October 9, 2015 in a limited release and through video on demand by Stage 6 Films and Vertical Entertainment. The film is directed by Todd Strauss-Schulson and […]
Machine learning models are getting quite good at generating realistic human faces — so good that I may never trust a machine, or human, to be real ever again. The new approach, from researchers at Nvidia, leapfrogs others by separating levels of detail in the faces and allowing them to be tweaked separately. The results are eerily realistic. The paper, published on preprint repository Arxiv (PDF), describes a new architecture for generating and blending images, particularly human faces, that “leads to better interpolation properties, and also better disentangles the latent factors of variation.” What that means, basically, is that the system is more aware of meaningful variation between images, and at a variety of scales to boot. The researchers’ older system might, for example, produce two “distinct” faces that were mostly the same except the ears of one are erased and the shirt is a different color. That’s not really distinctiveness — but the system doesn’t know that those are not important pieces of the image to focus on. It’s inspired by what’s called style transfer, in which the important stylistic aspects of, say, a painting, are extracted and applied to the creation of another image, which (if all goes well) ends up having a similar look. In this case, the “style” isn’t so much the brush strokes or color space, but the composition of the image (centered, looking left or right, etc) and the physical characteristics of the face (skin tone, freckles, hair). These features can have different scales, as well — at the fine side, it’s things like individual facial features; in the middle, it’s the general composition of the shot; at the largest scale, it’s things like overall coloration. Allowing the system to adjust all of them changes the whole image, while only adjusting a few might just change the color of someone’s hair, or just the presence of freckles or facial hair. In the image at the top, notice how completely the faces change, yet obvious markers of both the “source” and “style” are obviously present, for instance the blue shirts in the bottom row. In other cases things are made up out of whole cloth, like the kimono the kid in the very center seems to be wearing. Where’d that come from? Note that all this is totally variable, not just a A + B = C, but with all aspects of A and B present or absent depending on how the settings are tweaked. None of these are real people. But I wouldn’t look twice at most of these images if they were someone’s profile picture or the like. It’s kind of scary to think that we now have basically a face generator that can spit out perfectly normal looking humans all day long. Here are a few dozen: It’s not perfect, but it works. And not just for people. Cars, cats, landscapes — all this stuff more or less fits the same paradigm of small, medium, and large features that can be isolated and reproduced individually. An infinite cat generator sounds like a lot more fun to me, personally. The researchers also have published a new dataset of face data: 70,000 images of faces collected (with permission) from Flickr, aligned and cropped. They used Mechanical Turk to weed out statues, paintings, and other outliers. Given the standard dataset used by these types of projects is mostly red carpet photos of celebrities, this should provide a much more variable set of faces to work with. The dataset will be available for others to download here soon. ... Read More
A new study from Versus Systems and the MEMES (Management of Enterprise in Media, Entertainment & Sports) Center at UCLA’s Anderson School of Management examines at how gaming and advertising are evolving, and how one influences the other. As Versus Systems CEO Matthew Pierce put it, the goal was to study, “What is the impact on advertising as interactive media grows, and as more people consume interactive media?” The individual findings — People like rewards! Not everyone who plays games calls themselves a gamer! — may not be that shocking to TechCrunch readers. And since Versus Systems has built a white-label platform for publishers to offer in-game rewards, the study might also seem a bit self-serving. But again, this conducted was with UCLA’s Anderson School of Management, and both Pierce (who’s a lecturer at the school) and UCLA MEMES Head Jay Tucker pointed to size of the study, with 88,000 (U.S.-based) participants across a broad range of demographic groups. Of those respondents, 50 percent said that they’ve played a video game (on any platform) in the past week, while 41 percent said they’ve played a game in the past 24 hours. However, only 13 percent of respondents described themselves as gamers. That “identification gap” is even larger among women, where 56 percent played a game in the past week but only 11 percent identified themselves as gamers. Why does that matter? Well, the MEMES Center and Versus Systems argue in the study press release that “advertisers that are recognizing the value in advertising in-game may be underestimating how large and how diverse the gaming audience really is today.” How To Build A Smart Game Economy The study also suggests that traditional advertising may be facing more resistance from consumers, with 46 percent of respondents saying that they frequently or always avoid ads by “clicking the X” to close windows or changing channels or closing apps. Only 3.6 percent of respondents said they always watch ads all the way through. When asked what would make them play games more, the most popular answer was “winning real things that I want when I achieve things in-game” — it was the number one result for 30 percent of respondents, and among millennials, it did even better. (In comparison, 18 percent put “if the games were less expensive” as their top answer and 11 percent said “my friends playing the same game(s).”) This attitude even extended to TV, where 77 percent of respondents listed rewards as one of the things (not necessarily the top reason) that would make them watch more television. Meanwhile, 24 percent of respondents said listed “if more games/more shows were made for people like me” as the number one thing that would convince them to play or watch more. Tucker suggested that these seemingly scattershot answers are actually connected. On the advertising side, “We’ve got folks who are used to being part of a community all day, every day, whether that’s social media or massively multiplayer games. We see users are increasingly connected and are not really interested in getting pulled out of an experience. Rewards, if done properly, can reinforce being part of a community … you can amplify that sense of connection.” “The introduction of choice seems to make a big difference,” Pierce added. “We need new models where we can foster choice, foster community, foster more aspirational relationships between viewers and brands that ultimately allows content developers to have a relationship with the brands that isn’t so adversarial.” Meanwhile, when it comes to content and storytelling, Tucker said we’re entering an “age of personalization.” Among other things, that means more diversity, in what he described as “a generational shift away from stories that assume everybody’s looking at life from the same perspective.” Pierce and Tucker suggested that they’ll be taking an even closer look at the data in the coming months (“needs further study” was repeated several times during the interview), particularly by examining responses within smaller demographic groups. ... Read More
The best part of Nintendo’s Switch Online subscription service is the included classic games. Since launch, Nintendo has steadily added more NES titles, including some enhanced additions; just this week came a version of Metroid that starts you out with a huge array of powers just ahead of the big battle with Ridley. But what’s the best way to play these 8-bit classics? The Switch Joy-Con and Pro Controller are both passable, and there are also a few solid aftermarket controllers as well. Now, Nintendo has a different solution: a wireless recreation of the original NES gamepad. I’ve been testing them for a few days, and while they’re just about perfect for Switch Online NES games, there are a few hangups that will make you question the... Continue reading… ... Read More
CBS has released a new trailer for its upcoming second season of Star Trek: Discovery, highlighting the peril that will face the crews of the USS Discovery and USS Enterprise when the show returns in January. After the show teased his appearance in two earlier trailers, Ethan Peck’s bearded Spock is front and center, telling his half-sister, disgraced Starfleet officer Michael Burnham, that they’re facing a major challenge — one that could be a “defining moment for multiple civilizations, [and] billions of lives.” Their actions, he says, might determine the future of the galaxy — a fair statement, since Discovery is a prequel set before the original Star Trek TV series. This season’s main threat seems to be seven mysterious red bursts... Continue reading… ... Read More
Former EA Canada senior development director returns to games industry with new role at Sega's real-time-strategy studio... Read More
Peripheral maker asks users to install mining software in exchange for Razer Silver loyalty reward points... Read More
This is an area for widgets. You can add and edit widgets from the WordPress admin.
