J.C. van Gemert

Emotional valence categorization using holistic image features

Can a machine learn to perceive emotions as evoked by an artwork? Here we propose an emotion categorization system, trained by ground truth from psychology studies. The training data contains emotional valences scored by human subjects on the International Affective Picture System (IAPS), a standard emotion evoking image set in psychology. Our approach is based on the assessment of local image statistics which are learned per emotional category using support vector machines. We show results for our system on the I APS dataset, and for a collection of masterpieces. Although the results are preliminary, they demonstrate the potential of machines to elicit realistic emotions when considering masterpieces.

Episode-Constrained Cross-Validation in Video Concept Retrieval

Whereas video tells a narrative by a composition of shots, current video retrieval methods focus mainly on single shots. In retrieval performance estimation, similar shots in a narrative may result in performance overestimation. We propose an episode-based version of cross-validation leading up to 14% classification improvement over shot-based cross-validation.

Déjà vu: Motion Prediction in Static Images

This paper proposes motion prediction in single still images by learning it from a set of videos. The building assumption is that similar motion is characterized by similar appearance. The proposed method learns local motion patterns given a specific appearance and adds the predicted motion in a number of applications. This work (i) introduces a novel method to predict motion from appearance in a single static image, (ii) to that end, extends of the Structured Random Forest with regression derived from first principles, and (iii) shows the value of adding motion predictions in different tasks such as: weak frame-proposals containing unexpected events, action recognition, motion saliency. Illustrative results indicate that motion prediction is not only feasible, but also provides valuable information for a number of applications.

Interactive decision making using dissimilarity to visually represented prototypes

To make informed decisions, an expert has to reason with multi-dimensional, heterogeneous data and analysis results of these. Items in such datasets are typically represented by features. However, as argued in cognitive science, features do not yield an optimal space for human reasoning. In fact, humans tend to organize complex information in terms of prototypes or known cases rather than in absolute terms. When confronted with unknown data items, humans assess them in terms of similarity to these prototypical elements. Interestingly, an analogues similarity-to-prototype approach, where prototypes are taken from the data, has been successfully applied in machine learning. Combining such a machine learning approach with human prototypical reasoning in a Visual Analytics context requires to integrate similarity-based classification with interactive visualizations. To that end, the data prototypes should be visually represented to trigger direct associations to cases familiar to the domain experts. In this paper, we propose a set of highly interactive visualizations to explore data and classification results in terms of dissimilarities to visually represented prototypes. We argue that this approach not only supports human reasoning processes, but is also suitable to enhance understanding of heterogeneous data. The proposed framework is applied to a risk assessment case study in Forensic Psychiatry.

Accessing video archives using interactive search

In this presentation, we present a system for interactive search in video archives. In our view, interactive search is a four-step process composed of indexing, filtering, browsing, and ranking. We have experimentally verified, using 22 groups of two participants each, how users apply these steps in the interactive search and how well they perform.

Semantic Video Search

In this paper we describe the current performance of our MediaMill system as presented in the TRECVID 2006 benchmark for video search engines. The MediaMill team participated in two tasks: concept detection and search. For concept detection we use the MediaMill Challenge as experimental platform. The MediaMill Challenge divides the generic video indexing problem into a visual-only, textual- only, early fusion, late fusion, and combined analysis experiment. We provide a baseline implementation for each experiment together with baseline results. We extract image features, on global, regional, and keypoint level, which we combine with various supervised learners. A late fusion approach of visual-only analysis methods using geometric mean was our most successful run. With this run we conquer the Challenge baseline by more than 50%. Our concept detection experiments have resulted in the best score for three concepts: i.e. desert, flag us, and charts. What is more, using LSCOM annotations, our visual-only approach generalizes well to a set of 491 concept detectors. To handle such a large thesaurus in retrieval, an engine is developed which allows users to select relevant concept detectors based on interactive browsing using advanced visualizations. Similar to previous years our best interactive search runs yield top performance, ranking 2nd and 6th overall.

Making a Case for Learning Motion Representations with Phase

This work advocates Eulerian motion representation learning over the current standard Lagrangian optical flow model. Eulerian motion is well captured by using phase, as obtained by decomposing the image through a complex-steerable pyramid. We discuss the gain of Eulerian motion in a set of practical use cases: (i) action recognition, (ii) motion prediction in static images, (iii) motion transfer in static images and, (iv) motion transfer in video. For each task we motivate the phase-based direction and provide a possible approach.

Featureless: Bypassing feature extraction in action categorization

This method introduces an efficient manner of learning action categories without the need of feature estimation. The approach starts from low-level values, in a similar style to the successful CNN methods. However, rather than extracting general image features, we learn to predict specific video representations from raw video data. The benefit of such an approach is that at the same computational expense it can predict 2D video representations as well as 3D ones, based on motion. The proposed model relies on discriminative Wald-boost, which we enhance to a multiclass formulation for the purpose of learning video representations. The suitability of the proposed approach as well as its time efficiency are tested on the UCF11 action recognition dataset.

Per-patch metric learning for robust image matching

We propose a patch-specific metric learning method to improve matching performance of local descriptors. Existing methodologies typically focus on invariance, by completely considering, or completely disregarding all variations. We propose a metric learning method that is robust to only a range of variations. The ability to choose the level of robustness allows us to fine-tune the trade-off between invariance and discriminative power. We learn a distance metric for each patch independently by sampling from a set of relevant image transformations. These transformations give a-priori knowledge about the behavior of the query patch under the applied transformation in feature space. We learn the robust metric by either fully generating only the relevant range of transformations, or by a novel direct metric. The matching between query patch and data is performed with this new metric. Results on the ALOI dataset show that the proposed method improves performance of SIFT by 6.22% for geometric and 4.43% for photometric transformations.

Asymmetric kernel in Gaussian Processes for learning target variance

This work incorporates the multi-modality of the data distribution into a Gaussian Process regression model. We approach the problem from a discriminative perspective by learning, jointly over the training data, the target space variance in the neighborhood of a certain sample through metric learning. We start by using data centers rather than all training samples. Subsequently, each center selects an individualized kernel metric. This enables each center to adjust the kernel space in its vicinity in correspondence with the topology of the targets --- a multi-modal approach. We additionally add descriptiveness by allowing each center to learn a precision matrix. We demonstrate empirically the reliability of the model.