Imagine With Me

Imagine it for me: Generative Adversarial Approach for Zero-Shot Learning from Noisy Texts

Most existing zero-shot learning methods consider the problem as a visual semantic embedding one. Given the demonstrated capability of Generative Adversarial Networks(GANs) to generate images, we instead leverage GANs to imagine unseen categories from text descriptions and hence recognize novel classes with no examples being seen. Specifically, we propose a simple yet effective generative model...

Just Imagine

Program your NXT robot with Imagine

Our aim is to develop a versatile toolset for constructive learning. We achieve it through the marriage of powerful learning tools: a universal learning environment Imagine Logo (Kalaš and Hrušecká, 2004) and the recent and most popular educational robotics kit LEGO Mindstorms NXT. LEGO Mindstorms NXT is the second generation robotics toolkit. Performance and robustness have been improved, infr...

Imagine: An Interview with Svante Pääbo

Svante Pääbo works on the edge of what’s possible. He ignites our imagination, unlocking tightly held secrets in ancient remains. By patiently and meticulously working out techniques to extract genetic information from skin, teeth, bones, and excrement, Pääbo has become the leader of the ancient DNA pack. Sloths, cave bears, moas, wooly mammoths, extinct bees, and Neanderthals— all have succumb...

[Imagine this].