Where and What

where do we stand and what comes next?

Baryons: What, When and Where?

We review the current state of empirical knowledge of the total budget of baryonic matter in the Universe as observed since the epoch of reionization. Our summary examines on three milestone redshifts since the reionization of H in the IGM, z = 3, 1, and 0, with emphasis on the endpoints. We review the observational techniques used to discover and characterize the phases of baryons. In the spir...

What Went Where

We present a novel framework for motion segmentation that combines the concepts of layer-based methods and featurebased motion estimation. We estimate the initial correspondences by comparing vectors of filter outputs at interest points, from which we compute candidate scene relations via random sampling of minimal subsets of correspondences. We achieve a dense, piecewise smooth assignment of p...

Where-What Network 1: “Where” and “What” Assist Each Other Through Top-down Connections

This paper describes the design of a single learning network that integrates both object location (“where”) and object type (“what”), from images of learned objects in natural complex backgrounds. The in-place learning algorithm is used to develop the internal representation (including synaptic bottomup and top-down weights of every neuron) in the network, such that every neuron is responsible ...

Stacked What-Where Auto-encoders

We present a novel architecture, the “stacked what-where auto-encoders” (SWWAE), which integrates discriminative and generative pathways and provides a unified approach to supervised, semi-supervised and unsupervised learning without relying on sampling during training. An instantiation of SWWAE uses a convolutional net (Convnet) (LeCun et al. (1998)) to encode the input, and employs a deconvol...