Imran Saleemi

Probabilistic Modeling of Scene Dynamics for Applications in Visual Surveillance

We propose a novel method to model and learn the scene activity, observed by a static camera. The proposed model is very general and can be applied for solution of a variety of problems. The motion patterns of objects in the scene are modeled in the form of a multivariate nonparametric probability density function of spatiotemporal variables (object locations and transition times between them). Kernel Density Estimation is used to learn this model in a completely unsupervised fashion. Learning is accomplished by observing the trajectories of objects by a static camera over extended periods of time. It encodes the probabilistic nature of the behavior of moving objects in the scene and is useful for activity analysis applications, such as persistent tracking and anomalous motion detection. In addition, the model also captures salient scene features, such as the areas of occlusion and most likely paths. Once the model is learned, we use a unified Markov Chain Monte Carlo (MCMC)-based framework for generating the most likely paths in the scene, improving foreground detection, persistent labeling of objects during tracking, and deciding whether a given trajectory represents an anomaly to the observed motion patterns. Experiments with real-world videos are reported which validate the proposed approach.

Multi-source Multi-scale Counting in Extremely Dense Crowd Images

We propose to leverage multiple sources of information to compute an estimate of the number of individuals present in an extremely dense crowd visible in a single image. Due to problems including perspective, occlusion, clutter, and few pixels per person, counting by human detection in such images is almost impossible. Instead, our approach relies on multiple sources such as low confidence head detections, repetition of texture elements (using SIFT), and frequency-domain analysis to estimate counts, along with confidence associated with observing individuals, in an image region. Secondly, we employ a global consistency constraint on counts using Markov Random Field. This caters for disparity in counts in local neighborhoods and across scales. We tested our approach on a new dataset of fifty crowd images containing 64K annotated humans, with the head counts ranging from 94 to 4543. This is in stark contrast to datasets used for existing methods which contain not more than tens of individuals. We experimentally demonstrate the efficacy and reliability of the proposed approach by quantifying the counting performance.

Scene understanding by statistical modeling of motion patterns

We present a novel method for the discovery and statistical representation of motion patterns in a scene observed by a static camera. Related methods involving learning of patterns of activity rely on trajectories obtained from object detection and tracking systems, which are unreliable in complex scenes of crowded motion. We propose a mixture model representation of salient patterns of optical flow, and present an algorithm for learning these patterns from dense optical flow in a hierarchical, unsupervised fashion. Using low level cues of noisy optical flow, K-means is employed to initialize a Gaussian mixture model for temporally segmented clips of video. The components of this mixture are then filtered and instances of motion patterns are computed using a simple motion model, by linking components across space and time. Motion patterns are then initialized and membership of instances in different motion patterns is established by using KL divergence between mixture distributions of pattern instances. Finally, a pixel level representation of motion patterns is proposed by deriving conditional expectation of optical flow. Results of extensive experiments are presented for multiple surveillance sequences containing numerous patterns involving both pedestrian and vehicular traffic.

Discovering Motion Primitives for Unsupervised Grouping and One-Shot Learning of Human Actions, Gestures, and Expressions

This paper proposes a novel representation of articulated human actions and gestures and facial expressions. The main goals of the proposed approach are: 1) to enable recognition using very few examples, i.e., one or k-shot learning, and 2) meaningful organization of unlabeled datasets by unsupervised clustering. Our proposed representation is obtained by automatically discovering high-level subactions or motion primitives, by hierarchical clustering of observed optical flow in four-dimensional, spatial, and motion flow space. The completely unsupervised proposed method, in contrast to state-of-the-art representations like bag of video words, provides a meaningful representation conducive to visual interpretation and textual labeling. Each primitive action depicts an atomic subaction, like directional motion of limb or torso, and is represented by a mixture of four-dimensional Gaussian distributions. For one--shot and k-shot learning, the sequence of primitive labels discovered in a test video are labeled using KL divergence, and can then be represented as a string and matched against similar strings of training videos. The same sequence can also be collapsed into a histogram of primitives or be used to learn a Hidden Markov model to represent classes. We have performed extensive experiments on recognition by one and k-shot learning as well as unsupervised action clustering on six human actions and gesture datasets, a composite dataset, and a database of facial expressions. These experiments confirm the validity and discriminative nature of the proposed representation.

(MP) 2 T: multiple people multiple parts tracker

We present a method for multi-target tracking that exploits the persistence in detection of object parts. While the implicit representation and detection of body parts have recently been leveraged for improved human detection, ours is the first method that attempts to temporally constrain the location of human body parts with the express purpose of improving pedestrian tracking. We pose the problem of simultaneous tracking of multiple targets and their parts in a network flow optimization framework and show that parts of this network need to be optimized separately and iteratively, due to inter-dependencies of node and edge costs. Given potential detections of humans and their parts separately, an initial set of pedestrian tracklets is first obtained, followed by explicit tracking of human parts as constrained by initial human tracking. A merging step is then performed whereby we attempt to include part-only detections for which the entire human is not observable. This step employs a selective appearance model, which allows us to skip occluded parts in description of positive training samples. The result is high confidence, robust trajectories of pedestrians as well as their parts, which essentially constrain each others locations and associations, thus improving human tracking and parts detection. We test our algorithm on multiple real datasets and show that the proposed algorithm is an improvement over the state-of-the-art.

Learning semantic features for action recognition via diffusion maps

Efficient modeling of actions is critical for recognizing human actions. Recently, bag of video words (BoVW) representation, in which features computed around spatiotemporal interest points are quantized into video words based on their appearance similarity, has been widely and successfully explored. The performance of this representation however, is highly sensitive to two main factors: the granularity, and therefore, the size of vocabulary, and the space in which features and words are clustered, i.e., the distance measure between data points at different levels of the hierarchy. The goal of this paper is to propose a representation and learning framework that addresses both these limitations. We present a principled approach to learning a semantic vocabulary from a large amount of video words using Diffusion Maps embedding. As opposed to flat vocabularies used in traditional methods, we propose to exploit the hierarchical nature of feature vocabularies representative of human actions. Spatiotemporal features computed around interest points in videos form the lowest level of representation. Video words are then obtained by clustering those spatiotemporal features. Each video word is then represented by a vector of Pointwise Mutual Information (PMI) between that video word and training video clips, and is treated as a mid-level feature. At the highest level of the hierarchy, our goal is to further cluster the mid-level features, while exploiting semantically meaningful distance measures between them. We conjecture that the mid-level features produced by similar video sources (action classes) must lie on a certain manifold. To capture the relationship between these features, and retain it during clustering, we propose to use diffusion distance as a measure of similarity between them. The underlying idea is to embed the mid-level features into a lower-dimensional space, so as to construct a compact yet discriminative, high level vocabulary. Unlike some of the supervised vocabulary construction approaches and the unsupervised methods such as pLSA and LDA, Diffusion Maps can capture local relationship between the mid-level features on the manifold. We have tested our approach on diverse datasets and have obtained very promising results.

Multiframe Many---Many Point Correspondence for Vehicle Tracking in High Density Wide Area Aerial Videos

This paper presents a novel framework for tracking thousands of vehicles in high resolution, low frame rate, multiple camera aerial videos. The proposed algorithm avoids the pitfalls of global minimization of data association costs and instead maintains multiple object-centric associations for each track. Representation of object state in terms of many to many data associations per track is proposed and multiple novel constraints are introduced to make the association problem tractable while allowing sharing of detections among tracks. Weighted hypothetical measurements are introduced to better handle occlusions, mis-detections and split or merged detections. A two-frame differencing method is presented which performs simultaneous moving object detection in both. Two novel contextual constraints of vehicle following model, and discouragement of track intersection and merging are also proposed. Extensive experiments on challenging, ground truthed data sets are performed to show the feasibility and superiority of the proposed approach. Results of quantitative comparison with existing approaches are presented, and the efficacy of newly introduced constraints is experimentally established. The proposed algorithm performs better and faster than global, 1–1 data association methods.

Multimodal Analysis for Identification and Segmentation of Moving-Sounding Objects

In this paper, we propose a novel method that exploits correlation between audio-visual dynamics of a video to segment and localize objects that are the dominant source of audio. Our approach consists of a two-step spatiotemporal segmentation mechanism that relies on velocity and acceleration of moving objects as visual features. Each frame of the video is segmented into regions based on motion and appearance cues using the QuickShift algorithm, which are then clustered over time using K-means, so as to obtain a spatiotemporal video segmentation. The video is represented by motion features computed over individual segments. The Mel-Frequency Cepstral Coefficients (MFCC) of the audio signal, and their first order derivatives are exploited to represent audio. The proposed framework assumes there is a non-trivial correlation between these audio features and the velocity and acceleration of the moving and sounding objects. The canonical correlation analysis (CCA) is utilized to identify the moving objects which are most correlated to the audio signal. In addition to moving-sounding object identification, the same framework is also exploited to solve the problem of audio-video synchronization, and is used to aid interactive segmentation. We evaluate the performance of our proposed method on challenging videos. Our experiments demonstrate significant increase in performance over the state-of-the-art both qualitatively and quantitatively, and validate the feasibility and superiority of our approach.

Human Action Recognition across Datasets by Foreground-Weighted Histogram Decomposition

This paper attempts to address the problem of recognizing human actions while training and testing on distinct datasets, when test videos are neither labeled nor available during training. In this scenario, learning of a joint vocabulary, or domain transfer techniques are not applicable. We first explore reasons for poor classifier performance when tested on novel datasets, and quantify the effect of scene backgrounds on action representations and recognition. Using only the background features and partitioning of gist feature space, we show that the background scenes in recent datasets are quite discriminative and can be used classify an action with reasonable accuracy. We then propose a new process to obtain a measure of confidence in each pixel of the video being a foreground region, using motion, appearance, and saliency together in a 3D MRF based framework. We also propose multiple ways to exploit the foreground confidence: to improve bag-of-words vocabulary, histogram representation of a video, and a novel histogram decomposition based representation and kernel. We used these foreground confidences to recognize actions trained on one data set and test on a different data set. We have performed extensive experiments on several datasets that improve cross dataset recognition accuracy as compared to baseline methods.

Detection of independently moving objects in non-planar scenes via multi-frame monocular epipolar constraint

In this paper we present a novel approach for detection of independently moving foreground objects in non-planar scenes captured by a moving camera. We avoid the traditional assumptions that the stationary background of the scene is planar, or that it can be approximated by dominant single or multiple planes, or that the camera used to capture the video is orthographic. Instead we utilize a multiframe monocular epipolar constraint of camera motion derived for monocular moving cameras defined by an evolving epipolar plane between the moving camera center and 3D scene points. This constraint is parameterized as a polynomial function of time, and unlike repeated computations of inter-frame fundamental matrix, requires the estimation of fewer unknowns, and provides a more consistent separation between moving and static objects for different levels of noise. This constraint allows us to segment out moving objects in a general 3D scene where other approaches fail because their initial assumptions do not hold, and provides a natural way of fusing temporal information across multiple frames. We use a combination of optical flow and particle advection to capture all motion in the video across a number of frames, in the form of particle trajectories. We then apply the derived multi-frame epipolar constraint to these trajectories to determine which trajectories violate it, thus segmenting out the independently moving objects. We show superior results on a number of moving camera sequences observing non-planar scenes, where other methods fail.