Just to let you know, if you buy something featured here, Mashable might earn an affiliate commission. For some, jigsaw puzzles are a slow, relaxing pastime that exercise the creative and logic centers of the mind. If you're one of the former, it's time you experienced Clemens Habicht's 1,000 Colours. This is no ordinary jigsaw puzzle -- because of the subtle changes in color, it's a true test of patience, process, and attention to detail that's not for the faint of heart (or the faint of sight). You won't have the luxury of a real-world image to focus on while you try to match the picture on the box to the jumble of pieces in front of you.
The joy of any good jigsaw puzzle isn't finishing it, it's the satisfaction of linking pieces, one fit at a time. With the Infinite Galaxy Puzzle, which you can assemble in any direction and in countless shapes, that sensation need never end. Granted, that lack of resolution may make you crazy. But it makes the Infinite Galaxy Puzzle from Nervous System a unique contribution to the cannon. You'd expect nothing less from its creators, who have spent "five or six" years making jigsaw puzzles.
Eagen's strategy for completing the puzzles involves sorting the pieces by color into different paper plates. That way, if she's working on a section, she can sort through a pile of pieces that fit with that part of the image. Several puzzles on the walls come in the shape of maple leaves, and thus lack traditional starting points like edge and corner pieces.
Guys, it's been more than a year since we, as an office, started that three-thousand-piece puzzle depicting the Egyptian pyramids. But our "whoever has a few minutes and wants to put a few pieces together" mind-set has resulted in only a half-finished border and a few isolated pyramid peaks. I know I told you that several of our investors pulled their funding owing to our dismal first-quarter sales, but, truthfully, it's because of the puzzle. Every time they visit the office for a meeting, they notice how little progress we've made on the pyramids and it really bums them out. "If you can't put together an easy puzzle, you don't deserve my money," one of them yelled at me recently.
Generating music medleys is about finding an optimal permutation of a given set of music clips. Toward this goal, we propose a self-supervised learning task, called the music puzzle game, to train neural network models to learn the sequential patterns in music. In essence, such a game requires machines to correctly sort a few multisecond music fragments. In the training stage, we learn the model by sampling multiple non-overlapping fragment pairs from the same songs and seeking to predict whether a given pair is consecutive and is in the correct chronological order. For testing, we design a number of puzzle games with different difficulty levels, the most difficult one being music medley, which requiring sorting fragments from different songs. On the basis of state-of-the-art Siamese convolutional network, we propose an improved architecture that learns to embed frame-level similarity scores computed from the input fragment pairs to a common space, where fragment pairs in the correct order can be more easily identified. Our result shows that the resulting model, dubbed as the similarity embedding network (SEN), performs better than competing models across different games, including music jigsaw puzzle, music sequencing, and music medley. Example results can be found at our project website, https://remyhuang.github.io/DJnet.