USC digitally remasters Holocaust testimonial tapes

There are 4,754 interviews on damaged tapes, majority can be restored; new technology hopes to bring to life additional Holocaust testimony.

Holocaust candles 370 (photo credit: REUTERS)
Holocaust candles 370
(photo credit: REUTERS)
Eyewitness accounts of the Holocaust believed to be lost forever may soon be restored, according to experts at the University of Southern California’s Shoah Foundation.
The foundation, following the completion of a multi-year effort to digitize its archive of 52,000 Holocaust testimonies, is now beginning the process of digitally restoring the approximately 5 percent of tapes that it previously thought were irretrievable.
Between 1994 and 1999, the foundation’s staff interviewed tens of thousands of Holocaust survivors and eyewitnesses on analog tapes, which have since begun to degrade.
In January, after the foundation’s Information Technology Services department finished its efforts to digitize them, it began looking for ways to fix those tapes that it had not been able to digitize successfully.
“The 11,814 damaged tapes (4,754 interviews) have audio and/or visual problems that require additional fixing. 60 tapes (from 52 interviews) cannot be restored,” the foundation said in September.
The difficulty of digitizing the damaged tapes has led the foundation to develop new video restoration techniques, explained Ryan Fenton-Strauss, the foundation’s video archivist and post-production manager.
“It seemed terribly unfortunate that after a survivor had lived through the Holocaust and poured his or her heart into a testimony, that parts of it would be lost due to a technical problem during the recording process,” Fenton-Strauss said in a statement.
After looking through pictures of his daughter on photo-sharing service Picasa, he noticed that the site’s facial recognition software was able to detect his daughter’s face in different stages, from infancy until today.
“I realized then that if we could automate the process of identifying the ‘good’ and ‘bad’ images using image recognition software, then we could correct some of our most difficult video problems,” he said.
After tracking down more powerful facial recognition software, the archivist began the process of turning the videos into a series of still photos for processing.
A 30-minute video will “break down into 50,000 individual frames and 100,000 sub-frame fields,” Fenton- Strauss told technology blog Gizmodo. “Once you’ve got a series of pictures, you can identify the ones where there is no recognizable face. The first idea I had was if you simply drop all the stuff that’s bad and borrow the previous image; after filling it in, you would then sync back up with the audio and be able to watch the video. You’re sacrificing some resolution, essentially just isolating the good parts of the signal. By filling in the rest, it would play as a video.”
However, by using an advanced pattern recognition program, he was then able to “automate the search for patterns in bad images,” he said.
“There were survivors who managed to come out of the Holocaust and tell their story and, for some reason, it didn’t get captured,” he told the tech site. “It’s my job to make sure that we do the best we can for them – in fact, it’s an obligation, as we’ve been entrusted with these videos.”
This is not USC’s first foray into hi-tech approaches to the Holocaust.
In February the university announced that it would begin developing interactive, three-dimensional holograms of survivor testimonies. Participating survivors will sit in a room whose walls are covered with cameras and LED lighting and tell their personal stories while the cameras record them from all angles.
Later, the plan is to integrate the holograms with “natural language technology” similar to the iPhone’s personal assistant Siri or the Android platform’s Google Now, “which will allow people to engage with the testimonies conversationally by asking questions that trigger relevant, spoken responses,” according to the university’s Institute for Creative Technologies.