The DP of ‘Honey Boy’ Reveals Secrets from Her Big-Budget Hennessy Commercial

Want to shoot an atmospheric commercial in the style of The Neon Demon? This cinematographer can tell you how.

Natasha Braier is an Argentinian director of photography whose work includes visually striking feature films like Honey Boy, The Neon Demon, and The Rover. However, a commercial she shot for Hennessy X.O is as eerie and atmospheric as her previous work, and she mentions she’s taking her slightly creepy aesthetic from Neon Demon a step further here…which makes sense since the commercial was also directed by Nicolas Winding Refn.

Take a look at the Hennessy ad below:

CookeOpticsTV invited Braier to discuss the elements of this big-budget cognac commercial in a recent interview. Be warned, parts of the commercial are NSFW. Check it out below:

Shooting underwater

A portion of the commercial was shot with an actor underwater, with golden light being diffused through the water itself. She recalls the lights as Sputnik lights, probably on a tungsten setting.

Read More

The Best Filmmaking Deals of the Week (11.07.19)

Headlining our Deals of the Week, the Fujifilm X-T3 Mirrorless Camera just got a $200 price drop.

This week in filmmaking deals: Save $200 on not only the Fujifilm X-T3 Mirrorless Camera but also the NIKKOR Z 14-30mm f/4 S Lens. Also, you can get a $50 discount on both the Manfrotto Befree Travel Tripod and the Lowepro ProTactic Camera Backpack. Finally, get a whopping 47% discount on the Nebula 5100 3-Axis Handheld Stabilizer over at Adorama.

Read More

“This Little Live Show I was Doing on the Board”: DP Larry Sher on Joker

The Gotham City of Joker is a mere fraction of a degree removed from the New York City of 1981, a time and place Larry Sher knows well. The Hangover and Godzilla: King of the Monsters cinematographer grew up in nearby Teaneck, New Jersey and would sneak into the city on the bus as a teenager in the early 1980s. Sher channeled those experiences—as well as the seminal New York films of the era—to evoke the alienating urban nightmare of Gotham.  “My approach for Joker was to feed a little bit off of what the city looked like in my […]

Michael Beach Nichols Unmasks Wrinkles the Clown… or Does He?

The most important thing for Michael Beach Nichols was to tell the story honestly.

It would be easiest to say that the story of Wrinkles the Clown, the subject of Michael Beach Nichols’ new documentary, is actually a story wrapped in a rumor wrapped in a myth—with a real person at its true heart. The film describes the phenomenon of Wrinkles, a clown that Florida parents can call when they’re frustrated with their miscreant children to come to their houses and scare their children into obedience.

The idea is scary enough, having made its entrance by way of an ultra-creepy YouTube video of a clown crawling out from under a girl’s bed. What actually becomes truly frightening, as one watches the documentary, is the extent to which parents have embraced this method of scaring their children (the clown’s voicemail has 2-million-plus messages to date) and the seriousness with which some of the children take it.

Read More

Adobe’s #ProjectAboutFace can detect when portraits are altered and undo the edits

Adobe has collaborated with researchers at the University of California, Berkeley, to develop a new tool to detect when photos have been digitally altered using Photoshop’s Face-Aware Liquify tool and adjust them back to the original image.

The prototype tool, codenamed ‘Project About Face,’ will pore over an image pixel-by-pixel and create a heat map showing where it believes the image is most likely altered. From there, the edits can effectively be undone to show what the original image looked like.

According to Adobe, the tool is nearly twice as accurate as humans at detecting when a photograph has been altered. In its testing, Project About Face was able to detect altered images with 99-percent accuracy compared to the 53-percent accuracy of the human test group.

Since this only works with images edited inside Photoshop with its Face-Aware Liquify tool, the practical application isn’t widespread, but it’s a neat teaser nonetheless for future fake-detection methods.

Project About Face is just one of the many ‘Sneaks’ Adobe teased this year at Adobe MAX. Like many of the Sneaks, it’s unlikely we’ll see this tool available anytime soon, but it goes to show the growing possibilities of Adobe’s Sensei AI.

Leica SL2 – 47MP Full-Frame Sensor with 5K/30P and 4K/60P 10-Bit Video Internally

Leica has announced its new L-Mount full-frame mirrorless camera. The Leica SL2 features a 47MP full-frame CMOS sensor with 5-axis image stabilization. It is capable of recording 5K video at up to 30fps, 4K video at up to 60fps, and FullHD video at up to 180fps, all in 10-bit H.264. The camera features a high-resolution EVF, full-size HDMI port, two 3.5mm jack ports for headphones and microphone, and an all-metal magnesium body. It will be available from November 21st, 2019 for $5,995.

Leica SL2 full-frame mirrorless camera. Source: Leica

It has already been four years since the German premium brand introduced its Leica SL full-frame mirrorless camera. Our Leica SL review and Lab test articles from 2015 are a good reminder of its video capabilities. The newly announced Leica SL2 promises to bring even more video-oriented features. Let’s take a look!

Leica SL2 – 47MP Full-Frame Stabilized Sensor

The new Leica SL2 features a 47.3MP 3:2 full-frame CMOS sensor and Maestro III image processor, which enables up to 20 fps continuous stills shooting with an electronic shutter, 10 fps with a mechanical shutter, and 6 fps with a mechanical shutter and AF-C, for up to 78 consecutive DNG files with a color depth of 14 bits per RGB channel. The sensor features a 5-axis sensor-shift image stabilization system, which minimizes the appearance of camera shakes when shooting handheld.

Using sensor-shift stabilization technology, the SL2 features a Multishot mode to capture and compile eight separate exposures to create a single 187MP (16736 x 11168) RAW or JPEG file. This is suitable for static subjects and when working on a tripod.

Leica claims that the new SL2 offers a 14-stop Dynamic Range. The sensor’s design omits a low-pass filter for high degrees of sharpness and resolution. The ISO range is 50-50,000. Like its predecessor, the SL2 also uses Leica’s L-mount. Compared to 2015, however, thanks to the L-mount alliance, there are now much more affordable native lens choices from SIGMA and Panasonic.

Leica SL2 full-frame mirrorless camera. Source: Leica

The new Leica Object Detection AF, which uses 225 selectable AF areas for fast and accurate performance, ensures a fast and reliable autofocus. In addition to an array of modes and features, including face recognition, the SL2’s AF system also automatically detects whether a subject is stationary or in motion, and subsequently switches between motion and focus priority.

Leica SL2 Video Capabilities

We can really see that filmmakers are a very important segment for Leica. The SL2 seems to be a capable video camera with its 5K (4992 x 3744) 10-bit color depth video recording. The high-speed recording modes sound impressive as well – with UHD 4K60 (with 30 fps playback, 2x slow motion), and Full HD 180 fps (with 30 fps playback, 6x slow motion).

Leica SL2 full-frame mirrorless camera. Source: Leica

The list of internal recording modes looks impressive, too. All of the following modes are in H.264 with 10-bit color depth, with MOV quicktime codec (UHD and Full HD are also available in MP4):

  • 5K (4992 x 3744) at 23.976p/25p/29.97p
  • DCI 4K (4096 x 2160) at 24.00p/25p/29.97p/50p/59.94p
  • UHD 4K (3840 x 2160) at 23.976p/25p/29.97p/50p/59.94p
  • Full HD (1920 x 1080) at 23.976p/25p/29.97p/50p/59.94p/100p/119.88p/180p

The camera, unfortunately, has a recording limit – up to 29 minutes. We don’t know yet whether the SL2 uses 4:2:2 or 4:2:0 subsampling and what the exact bitrates for the recording modes are. We will update the article once we get this information.

Leica SL2 main display. Source: Leica

There is a “Cine mode”, which (when activated) transforms the SL2 into a manually controlled cine camera as ISO becomes ASA, the shutter speed is marked in degrees on the rotary disk shutter, and the f-stops indicating the aperture ratio are replaced by T-stops, which measure the actual amount of light transmitted through the lens.

The camera offers a dedicated Hybrid Log Gamma (HLG) Photo Style mode and L-Log gamma profile for higher Dynamic Range and greater flexibility for color grading. The Leica SL2 also outputs video signal (in DCI 4K (4096 x 2160) 10-bit up to 59.94p) over its full-size HDMI port, so using an external monitor or recording to an optional external recorder is possible. Finally, there are also dedicated 3.5mm jack ports for a microphone and headphones.

Leica SL2 Ergonomy

Another great feature is a high resolution electronic viewfinder – the Leica SL2 includes a new 5.76m-dot EyeRes OLED EVF with 0.78x configurable viewing magnification, along with 120 fps refresh rate, 0.005 sec lag, and 10,000:1 contrast ratio. On the back, there is a large 3.2″ 2.1m-dot touchscreen LCD with durable Gorilla glass. Last but not least, on the top there is a 1.28″ top status LCD, which can be used for quick recognition of shooting settings – even when the camera is powered off.

Leica SL2 top display. Source: Leica

Leica SL2’s durable, weather-sealed magnesium-alloy all-metal body is splash-, dust-, and freeze-proof with an IP54 certification. The SL2 is actually the only mirrorless camera designed and manufactured in Germany. The BP-SCL4 rechargeable lithium-ion battery is rated for approximately 370 shots per charge when working with the rear monitor.

The camera offers dual UHS-II-compatible SD card slots. The USB Type-C port allows for fast file transferring, as well as high-speed in-camera battery charging. Additionally, built-in Wi-Fi and Bluetooth work in conjunction with the Leica FOTOS app (iOS and Android) for wireless file transferring and remote camera control.

Leica HG-SCL6 Battery Grip

Along with the launch of the Leica SL2 camera, the German camera manufacturer also announced a new battery grip for this camera. As one would expect, the HG-SCL6 grip offers both, extended battery life and a more comfortable grip when shooting in vertical orientation. It accepts one additional BP-SCL4 battery to effectively double the overall battery life. The grip’s design maintains the same layout, fit, and finish as the camera body itself for seamless use, and incorporates a secondary shutter button and control dials.

Leica HG-SCL6 battery grip. Source: Leica

Price and Availability

The Leica SL2 will available from November 21st 2019 on. It can already be pre-ordered now. The price for the camera has been set to be $5,995.

I suppose the battery grip will be available at the same time. It is priced at $995 and can be pre-ordered now, as well.

What do you think of the new Leica SL2? Have you worked with the original Leica SL? Let us know in the comments underneath the article.

The post Leica SL2 – 47MP Full-Frame Sensor with 5K/30P and 4K/60P 10-Bit Video Internally appeared first on cinema5D.

SIGMA 24-70mm F/2.8 Art Lens Announced – for Sony E-Mount and L-Mount Full-Frame Cameras

SIGMA has just announced their next Art zoom dedicated for mirrorless cameras: the SIGMA 24-70mm F/2.8 DG DN Art. This new flagship lens is available for E-mount (Sony) and L-mount (Panasonic, Leica, and SIGMA) cameras. Let’s take a closer look!

Sigma2870DGDN_02

Image credit: SIGMA

SIGMA 24–70mm F/2.8 DG DN Art

Usually, there are three zoom lenses in every professional’s toolkit to cover nearly every shooting scenario: a wide-angle zoom, a standard zoom, and a telephoto zoom lens. For professional use, all these lenses need to have a fast, constant aperture, usually at F/2.8.

After the 14-24mm F/2.8 DG DN Art that launched last year, the new SIGMA 24–70mm F/2.8 DG DN Art is the flagship standard zoom lens of the new-generation Art zoom.

Sigma2870DGDN_01

Image credit: SIGMA

These SIGMA zoom lenses are “DG,” which means they are designed to deliver the ultimate in performance on cameras with full-frame sensors. The “DN” in the name indicates that these lenses are for mirrorless cameras. Indeed, the SIGMA 24–70mm F/2.8 DG DN Art is available in Sony E-mount and L-mount.

Sigma2870DGDN_03

The SIGMA 24-70mm F/2.8 DG DN Art with the SIGMA FP camera. Image credit: SIGMA

SIGMA 24–70mm F/2.8 DG DN Art Features

The name of the lens also makes clear: there is no “OS” in it, so it doesn’t feature image stabilization. The lack of lens stabilization is a pretty wise choice from SIGMA, because it allows them to reduce the overall weight/size/price of the 24-70mm F/2.8 DG DN Art. Also, most of the Full-Frame mirrorless cameras features in-body image stabilization. As a result, the lens only weighs 835 grams, whereas the DSLR version of the lens weighs 1025 grams.

The lens design consists of 19 elements in 15 groups. There are six FLD elements (low dispersion glass) and two SLD elements (special low dispersion). Also, there are three aspheric lenses to reduce axial chromatic aberration and sagittal coma aberration. There is a Super Multi-Layer Coating on the optics, as well as a Nano Porous Coating to reduce unwanted lens flares. The aperture diaphragm consists of 11 rounded blades. The minimum focusing distance is 18 cm at the wide-angle end.

Sigma2870DGDN_04

Image credit: SIGMA

The lens is dust- and splash-proof, and features a zoom lock switch to prevent the lens barrel from extending unexpectedly.

Pricing and Availability

The SIGMA 24-70mm F/2.8 DG DN Art comes with a lens hood. It will be available mid-November this year. There is no information on pricing yet.

What do you think of the SIGMA 24-70mm F/2.8 DG DN Art? Do you think it pairs nicely with the SIGMA fp camera? Let us know in the comments!

The post SIGMA 24-70mm F/2.8 Art Lens Announced – for Sony E-Mount and L-Mount Full-Frame Cameras appeared first on cinema5D.

5 Common Portrait Photography Mistakes and How to Fix Them

Portrait photographer Miguel Quiles is back with another useful video for aspiring portrait photographers. This time, he’s put together a video outlining the five most common portrait photography mistakes that he’s seen over the years, and how he suggests that beginners avoid or fix them.

Quiles tells PetaPixel that the list is based on his personal experience “doing portfolio reviews or judging portraits in competitions.” So while these may seem pretty “basic,” they’re also incredibly common across the experience spectrum. The five mistakes are:

  1. Missing Focus – Don’t wait until after your shoot to zoom in to 100% and realize you missed critical focus
  2. Poor Composition – Your viewer shouldn’t have to struggle to figure out where their eyes should go
  3. Photoshop Gymnastics – Don’t spend hours trying to fix a bad image in post; take it as a lesson that you should address poor lighting, bad styling, and other issues while shooting whenever possible.
  4. Adding Watermarks – Quiles claims that you’ll be hard-pressed to find top photographers who use massive, distracting watermarks. He suggests embedding your info into metadata instead.
  5. Over-Retouching – Overuse of things like Frequency Separation, Dodge and Burn, and others are some of the most common and easiest ways to ruin a great portrait.

Check out the full video to hear Quiles’ thoughts on each of these five points, how he sees these mistakes pop up most often, and what he suggests you do to avoid or fix them. Some, like the Watermarking tip, will no doubt stir up some controversy, but there’s a reason behind each of the points that he makes.

FANTASTIC BEASTS – THE CRIMES OF GRINDELWALD: VFX Breakdown by Rodeo FX

Rodeo FX have released the VFX Breakdown about their work on FANTASTIC BEASTS – THE CRIMES OF GRINDELWALD:

WANT TO KNOW MORE?
Rodeo FX: Dedicated page about FANTASTIC BEASTS – THE CRIMES OF GRINDELWALD on Rodeo FX website.

© Vincent Frei – The Art of VFX – 2019

The post FANTASTIC BEASTS – THE CRIMES OF GRINDELWALD: VFX Breakdown by Rodeo FX appeared first on The Art of VFX.

Elle Germany Apologizes for Implying that Black Models are a Fashion Trend

Elle Germany has issued a public apology after the magazine’s most recent issue declared that “Black is Back,” a statement that has been widely interpreted as implying that black models are a “trend” to be capitalized on by the fashion industry.

The controversy first broke on October 29th, when popular Instagram account Diet Prada pointed out the offending page, as well as several other mistakes they found in the November 2019 “Back to Black” issue of the magazine. Not only was the phrase “Black is Back” problematic, Diet Prada pointed out that model Naomi Chin Wing was misidentified as Janaye Furman, and was further incredulous that the “Back to Black” issue featured a white model on the cover.

“For their November 2019 issue, the presumably white-led publication declares that ‘black is back.’ Ironic when they, along with much of the fashion industry, have been complicit in denying visibility to black models until relatively recently,” reads the post’s caption. “The issue, titled ‘Back to Black,’ also features a white model on the cover. You can’t make this stuff up!”

View this post on Instagram

Not a good look, @ellegermany . For their November 2019 issue, the presumably white-led publication declares that “black is back”. Ironic when they, along with much of the fashion industry, have been complicit in denying visibility to black models until relatively recently. Oh, and apparently they can’t actually tell models apart. In the bottom middle, a picture of @naomichinwing is used in place of @iam_janaye . And @joansmalls has been around for a minute/hasn’t gone anywhere lmao. The issue, titled “Back to Black”, also features a white model on the cover. You can’t make this stuff up! • #elle #ellemagazine #ellegermany #naomichinwing #janayefurman #models #blackmodels #modelsofcolor #runway #fashionweek #fashionmonth #pfw #nyfw #mfw #lfw #paris #london #milan #nyc #wtf #fail #magazine #print #editorial #editor #editorinchief #media #sabinenedelchev #dietprada

A post shared by Diet Prada ™ (@diet_prada) on

The response to Diet Prada’s Instagram post, which has been liked over 98,000 times as of this writing, was swift and furious. At best, commenters referred to the the gaff as “good intention, poor execution”; at worst, the page and wording was dubbed “sickening,” “racist,” “offensive on every level,” and “the cover line from hell.”

According to TODAY, fashion icon Naomi Campbell even joined the conversation, releasing a statement that called the mistake “highly insulting in every way” and offering to sit down with Elle Germany and offer some guidance on cultural sensitivity. “I’ve said countless of times we are not a TREND,” said Campbell.”We are here to STAY. It’s ok to celebrate models of color but please do it in an ELEGANT and RESPECTFUL way.”

The backlash quickly made its way back to Elle, and the publication issued a public apology on its own Instagram account less than a day later.

“In our current issue, we are approaching the color black from different angles. As one of our topics, it was our aim to feature strong black women who work as models for the fashion industry. In doing so, we have made several mistakes for which we apologize to anyone we might have hurt,” reads the statement.

“It was a mistake to use the cover line ‘Back to black’ which could be understood as if black persons would be a kind of fashion trend. This obviously wasn’t our intention and it was our mistake not to be more sensitive about this.”

(via TODAY via Fstoppers)

The Kandao QooCam 8K is an affordable pocket-sized 8K 360-degree camera

360-degree cameras cover a much larger field of view than conventional cameras, which is why a high pixel count is essential for capturing good detail. Most current consumer models max out at 4K resolution, though, with 8K video reserved for bulkier high-end models that are mostly aimed at professionals.

Chinese manufacturer Kandao is now changing this, however. Today, the company has launched QooCam 8K, which it calls the ‘first pocketable 8K 360-degree camera.’

The camera uses a pair of fisheye lenses and 1/1.7” sensors to capture stills and video; it’s also capable of 12bit Raw capture. It offers automated image stacking at Raw level for increased dynamic range and detail and can also record 4K video at 120 frames per second.

Kandao’s Super Steady electronic stabilization uses a 6-axis gyro to smooth video footage without the need for a gimbal and, according to the manufacturer, makes the QooCam 8K usable as an action cam.

A 2.4-inch OLED touchscreen lets you check shooting information, preview and playback footage and adjust parameters without connecting a smartphone. Footage can be edited and reframed in the dedicated QooCam App or QooCam Studio which include templates for those who are beginners in the world of 360-degree editing.

Additionally, the QooCam 8K can be used to live stream 360-degree video in 4K with in-camera real-time stitching. It is compatible Facebook, Youtube and other platforms. The
Kandao QooCam 8K 360-degree camera is now available for pre-order on the Kandao website for 599 Euros (approximately $660). Shipping is scheduled for December 2019.

Xiaomi patent hints at future smartphone with pop-up selfie camera and 8K recording

A new patent belonging to Chinese company Xiaomi has surfaced revealing a new camera design that includes a pop-up selfie camera and a rear square zoom lens. The patent was first spotted by Dutch website Let’s Go Digital, which notes that the patent was recently registered by the Hague International Design System.

The Xiaomi patent covers a mobile phone that features a vertical array of cameras on the back of the handset; the top square lens is believed to possibly be a zoom lens, whereas the lower three circles potentially include another camera, an LED flash, and a third sensor of some type.

The pop-up illustration below shows how the company will handle offering a front-facing camera that doesn’t blemish the model’s bezel-less display.

The patent was granted on August 9, about a month before XDA Developers revealed the existence of a new Xiaomi smartphone featuring a camera with 5x optical zoom and 50x digital zoom. The discovery was made in the beta version of a Xiaomi software update, which itself came shortly before the official unveiling of its new Mi Note 10.

The Mi Note 10 features five cameras, including one with 50x digital zoom and one with 5x optical zoom. It’s unclear whether XDA’s discovery was ultimately pointing toward the Mi Note 10 or a future Xiaomi model, which may have been teased in the newly-published mobile phone patent.

The details uncovered in the software update indicated that the related Xiaomi phone would include 8K/30fps video recording capabilities. In that case, the combination of a newly granted patent and the beta software details make the case for there being another major Xiaomi smartphone in the pipeline, one that may be introduced in 2020.

Indie Filmmakers Have a Special Tube Light Option with the Nanlight PavoTube

Nanlite now has an indie-friendly lighting tube option available with the Nanlite PavoTube.

RGB Lighting Tubes have been everywhere in the last few years. You can’t watch a music video or an action movie without seeing them both in the frame and in the behind the scenes videos.

Why?

Well, a light tube gives a nice, soft source that is easy to build into a set, or hide near one, in a simple package.

Read More

Getty Images is Going All Royalty Free, and That Sucks for Photographers

Getty Images has announced that it will be transitioning all of its creative stock photography offerings to a Royalty Free licensing model and ditching Rights Managed licensing entirely—a move that will be more convenient for customers, but worse for photographers.

The news was announced via email, and from a business perspective it makes sense. Getty claims that it’s seen a “steady […] year‑over‑year decline in Creative RM à la carte licenses over the last five years, with declines accelerating over time.” While the company doesn’t share any data to back up this claim, we believe it, as almost all of the newer paid and free stock photography platforms on the market have embraced the one-size-fits-all Royalty Free licensing model.

“Over the years, customers’ needs have changed,” reads the announcement email. “Complicated licensing models create friction and customers demand simplicity—they want the most simple and most flexible access to relevant, authentic imagery.”

But while Getty goes to great lengths to point out how great this move to RF-only will be for “buyers” and “customers,” the announcement all-but-ignores the other major player in this trio: the photographers.

Beyond telling photographers when their Rights Managed images will disappear from single-use image search (by the end of January 2020) and explaining how they’ll be allowed to use or re-submit those images once they’ve been removed, Getty only mentions the interests of photographers once in the entire email.

“This will benefit customers,” writes Getty… again, “and provide an opportunity to grow overall licensing volume and revenue for both Getty Images and our contributors.”

While the volume of licenses purchased may indeed increase, the fact remains that contributors will lose control over how their licensed images are used. If you want to sell a stock photograph through Getty—the largest player in the game—you will no longer be able to dictate pricing based on use-case, and the option for exclusivity disappears from the page entirely. The only options available will be sizing:

Unfortunately, this move was more-or-less inevitable and probably won’t change much for the photographers who are already selling their stock photography through Getty Images and iStock. Royalty Free is already the default Filter used when searching for imagery on Getty, and as the announcement points out, it’s also “the preferred and dominant licensing model for our customers ­due to the simplicity, value and quality available.”

It’s just one more way that photographers are being asked to trade control for exposure to the largest possible audience of potential buyers.

To learn more about this transition, you can read the full announcement email here. According to this, the RM model should be completely “retired” by November 2020, and RM images will be removed from single image licensing by the end of January 2020. This change applies to Creative image licensing only, and will not impact Rights Managed editorial stills licensing offered by Getty.

(via DPReview)

SMPTE Hollywood explores Deepfakes: creative promise and potential perils

SMPTE Hollywood explores Deepfakes and their use for nefarious purposes

The next SMPTE Hollywood Section meeting explores “deepfakes”, and how the emerging field of synthetic humans can be used for both entertainment and nefarious purposes.

Under the title “Digital Humans and Deepfakes: Creative Promise and Peril”, the next SMPTE Hollywood Section meeting promises to be a memorable moment everyone should attend. The Hollywood Section of SMPTE, the organization defining the future of storytelling, will examine the promise, and potential peril, of digital humans and so-called deepfakes at its monthly meeting on Tuesday, November 19, in Hollywood. Held in tandem with the Radio, Television, Digital Newsroom Association (RTDNA), the free event will include a panel discussion of experts in the emerging field of synthetic humans.

Deepfakes are believable human images synthesized through artificial intelligence techniques from completely real and totally non-real or “fake” elements. In Hollywood, digital humans, convincing enough to fool audiences, have been the holy grail of visual effects for decades. Both techniques are designed to trick the viewer, but, whereas digital humans are constructed to entertain, deepfakes can be used to mislead and misinform, often for non entertainment purposes.

SMPTE Hollywood explores Deepfakes and their use for nefarious purposes

Deepfake videos doubled the last 9 months

Everywhere you look online these days, deepfakes are present. According to an article on BBC News, “new research shows an alarming surge in the creation of so-called deepfake videos, with the number online almost doubling in the last nine months. There is also evidence that production of these videos is becoming a lucrative business.”

The trend started, apparently, when someone, back in December 2017, used deep learning technology to edit the faces of celebrities onto people in pornographic video clips. The term is used for both the technology used and the videos created, which can be anything from a funny transformation to a video used for political propaganda. The multiple examples shown in recent months reveal how far the technology has evolved, and how dangerous it can become if used for nefarious purposes.

We’re on the verge of a revolution that will mix Artificial Intelligence, machine learning and immersive experiences made possible by Augmented Reality and Virtual Reality, to take us to what can well be the uncanny valley. The deepfake experiences we’ve seen until now are mostly bidimensional, but when you add VR to the equation, you get a multidimensional experience that enhances the illusion. The “holographic” promise of Star Trek may be just around the corner. I recently met Mona Lisa, smiling and waving at me,  just by using my Oculus Rift S headset…

From left to right, Christopher Nichols, Niko Pueringer, Shruti Agarwal and Debra Kaufman

Digital humans and deepfakes

The movie-industry, that thrives on a make-believe world, is now worried about the potential peril represented by “deepfakes”. SMPTE Hollywood and the RTDNA will offer a totally real presentation on deepfakes and digital humans. The panel will describe the history of digital humans and deepfakes, the challenges involved in creating them convincingly, and if/how news and entertainment professionals can spot a deepfake.

 

Presenters include Chaos Group Lab head of research and development Christopher Nichols, who leads the Digital Human League, sponsor of the open source Wikihuman; Corridor Digital’s Niko Pueringer, who has produced short-form Internet content for more than a decade and is an expert in creating and detecting deepfakes, and Shruti Agarwal, a Ph.D. student in the Department of Electrical Engineering and Computer Sciences at University of California, Berkeley, who is conducting research in multimedia forensics.

Freelance journalist Debra Kaufman (USC Entertainment Technology Center, New York Times, Los Angeles Times, Wired, Reuters, Bloomberg American Cinematographer, International Cinematographers Guild Magazine) will moderate the discussion. Kaufman and Linda Rosner are producers of the event.

Taxi Driver starring Al Pacino

To illustrate this piece of news, we picked a few examples that are directly connected to Hollywood, from Taxi Driver starring Al Pacino to Terminator 2 starring Sylvester Stallone, or Willem Dafoe in the Coin Toss sequence in No Country for Old Men.  These deepfake videos, created by Patreon author named ctrl shift face, are a good example of what is possible to achieve right now with the technology. The author is “creating entertaining deepfake videos”, which he says are “windows to parallel universes”. The videos and the behind-the-scenes on some of them might help to educate people about what “deepfakes”, but they are a frightening sign of how easy it can be to fool people… Taxi Driver with Al Pacino!?

The SMPTE Hollywood Section, November meeting takes place Tuesday, November 19, 2019. 6:30 p.m. at Hollywood American Legion Post 43, 2035 N. Highland Ave., Los Angeles, CA 90068. The title is “Digital Humans and Deepfakes: Creative Promise and Peril”. Follow the link to register.

Early arrivals have an extra activity: a guided tour of the historic American Legion #43, built in 1929. Tours will include the modern projection booth with 35/70 mm FILM and featuring Christie 4K DIGITAL Projection. Please indicate on EVENTBRITE your choice of a tour at 5, 5:30 or 6 pm.  As always, SMPTE Hollywood Section meetings are FREE and non-members are welcome.

The post SMPTE Hollywood explores Deepfakes: creative promise and potential perils appeared first on ProVideo Coalition.

Take More Photos, Spend Less Money

Photography can be an expensive pursuit, and the cost of things (and the pressure to buy them all, buy them new, and buy them now) can get in the way of putting that money in better places.

It’s not my place to tell you what to do with your money, but I don’t want to suggest you consider investing it in ways that give you greater creative freedom and make you a better photographer than the usual push you might be getting to buy the latest and greatest, which I’ve found to be an endless pursuit.

Here are four recommendations I want to nudge you to consider, none of them particularly novel, but I thought it might just help to hear it from me.

1. Resist the upgrade cycles. I know, I know, New Camera MK VII just got released and it’s sexy and shiny but it won’t make your photographs any better. Not these days. Unless you really do need ISO 250,000 for the series you’re doing on deep-earth spelunking. And the second version of that lens that you loved only months ago? You don’t need it. Skip an upgrade cycle. Let others pay top dollar to fund the R&D on the newest and shiniest. In a year, it’ll be obsolesced by yet another development. Buy it then. Or don’t. You don’t need the pro bodies, the L-glass, or the red dot to be a “real” photographer.

2. Don’t buy at all. Not everyone gets to do this but if you’re combining your craft with commerce and you’re making even just part of your living with photography, don’t buy the fancy lens you need for the gig; rent it and bill it to the client.

Technology is a terrible place to put your money, and the less overhead you have in gear, the better. Put that $2000 into your registered retirement savings (RRSP/401K) instead. I know, it’s less sexy, but it’s the better move. You get to keep $2000, avoid the interest on the credit card you’d inevitably have to use to buy the lens in the first place, and you get a tax break. And if you’re not doing this professionally, renting can still save you money. As can borrowing. You don’t have to own it to enjoy it and test it out.

3. Still have to buy? Consider buying it used. Most of us are really precious about our gear and we want it to be perfect. But my gear, perfect as it often is when I buy it, no longer is. Now it’s just my dented up, scratched-up gear that still works for me day in and day out, and if it’s going to be a little dinged up (but still working perfectly) in a year, you might as well buy it pre-dinged and save the money.

There are some great ways to buy used gear, often at a fraction of the original cost. Consider your local camera store first; they often have customers trading in gently-used gear for new gear, and they’ll stand behind what they sell. The same applies to some of the giants of photography retail, like B&H in New York. KEH.com (also based in the US) is very reliable for both selling and buying.

4. Sell It. Finally, if it’s not being used, consider selling it. The longer you wait, the less your gear will be worth. And if it’s not being used, it could be traded in for something that you will use—something that you might otherwise spend more good money on. Unless you’re an actual collector, there is usually no value in keeping that gear around; most often, the value diminishes the longer you wait.


Having gone bankrupt once in my life, I’m full of some very passionate sermons about money, and this can be a very expensive craft. It doesn’t have to be.


So let me wrap it up with one last plea: if you’re going to spend your money, spend it where it really matters.

  • Spend it on opportunities to learn, to shoot more, and to travel (if that’s your thing).
  • Take a workshop or rent a studio for a week and work on your project.
  • Spend it on a printer or on getting prints made so you can finally hold your work in your hands.
  • Spend it on books of great photographs and study them.

Be as intentional about your money as you are about your photographs and you’ll have more opportunities to make those photographs—and if you’re wise about where you spend it, you’ll become a stronger photographer at the same time.


About the author: David duChemin is a world & humanitarian assignment photographer, best-selling author, international workshop leader, and host of the podcast A Beautiful Anarchy. You can find out more about him and see more of his work on his website or by following him on Instagram and YouTube. This post was also published here.

Video Captures the Moment a Woman Almost Fell Into the Grand Canyon While Taking a Photo

An onlooker at the Grand Canyon in Arizona captured the harrowing instant when a visitor almost fell into the canyon while taking a photo of her mom. The video of the incident has since gone viral: a perfect example of what not to do.

The footage shows 20-year-old Emily Koford backing up, trying to compose the perfect photo of her mom Erin, when she takes one step too many. The footage was captured by Kevin Fox, who saw the duo posing and started filming in order to show his kids “the stuff you don’t do.”

“They’re walking around and I think, this doesn’t look good,” Kevin Fox told ABC News last Friday. “As she starts walking backwards, I just gasp.”

Once the video went viral, ABC caught up with Koford and her mother as well, both of whom feel understandably lucky that this didn’t turn out much worse:

According to ABC’s report, there have been 64 total deaths at the park since it first opened, and while heat stroke and other causes are more common than falling off the edge, a spokesperson for the park told ABC that two to three people die each year from incidents just like this one.

Fortunately for Koford, she was able to catch herself on a rock, turning this story into a cautionary tale rather than a tragedy. Others before her have not been so lucky.

(via ABC News via Fstoppers)