arXiv subjects

Peter Robinson

Publications and source records attributed to Peter Robinson.

69 records · Page 4Linked to original sources

arXiv2015

An Empirical Study of Recent Face Alignment Methods

The problem of face alignment has been intensively studied in the past years. A large number of novel methods have been proposed and reported very good performance on benchmark dataset such as 300W. However, the differences in the experimental setting and evaluation metric, missing details in the description of the methods make it hard to reproduce the results reported and evaluate the relative merits. For instance, most recent face alignment methods are built on top of face detection but from different face detectors. In this paper, we carry out a rigorous evaluation of these methods by making the following contributions: 1) we proposes a new evaluation metric for face alignment on a set of images, i.e., area under error distribution curve within a threshold, AUC$_α$, given the fact that the traditional evaluation measure (mean error) is very sensitive to big alignment error. 2) we extend the 300W database with more practical face detections to make fair comparison possible. 3) we carry out face alignment sensitivity analysis w.r.t. face detection, on both synthetic and real data, using both off-the-shelf and re-retrained models. 4) we study factors that are particularly important to achieve good performance and provide suggestions for practical applications. Most of the conclusions drawn from our comparative analysis cannot be inferred from the original publications.

cs.CV↗

arXiv2015

Human and Sheep Facial Landmarks Localisation by Triplet Interpolated Features

In this paper we present a method for localisation of facial landmarks on human and sheep. We introduce a new feature extraction scheme called triplet-interpolated feature used at each iteration of the cascaded shape regression framework. It is able to extract features from similar semantic location given an estimated shape, even when head pose variations are large and the facial landmarks are very sparsely distributed. Furthermore, we study the impact of training data imbalance on model performance and propose a training sample augmentation scheme that produces more initialisations for training samples from the minority. More specifically, the augmentation number for a training sample is made to be negatively correlated to the value of the fitted probability density function at the sample's position. We evaluate the proposed scheme on both human and sheep facial landmarks localisation. On the benchmark 300w human face dataset, we demonstrate the benefits of our proposed methods and show very competitive performance when comparing to other methods. On a newly created sheep face dataset, we get very good performance despite the fact that we only have a limited number of training samples and a set of sparse landmarks are annotated.

cs.CV↗

arXiv2015

Face Alignment Assisted by Head Pose Estimation

In this paper we propose a supervised initialization scheme for cascaded face alignment based on explicit head pose estimation. We first investigate the failure cases of most state of the art face alignment approaches and observe that these failures often share one common global property, i.e. the head pose variation is usually large. Inspired by this, we propose a deep convolutional network model for reliable and accurate head pose estimation. Instead of using a mean face shape, or randomly selected shapes for cascaded face alignment initialisation, we propose two schemes for generating initialisation: the first one relies on projecting a mean 3D face shape (represented by 3D facial landmarks) onto 2D image under the estimated head pose; the second one searches nearest neighbour shapes from the training set according to head pose distance. By doing so, the initialisation gets closer to the actual shape, which enhances the possibility of convergence and in turn improves the face alignment performance. We demonstrate the proposed method on the benchmark 300W dataset and show very competitive performance in both head pose estimation and face alignment.

cs.CV↗

arXiv2015

Rendering of Eyes for Eye-Shape Registration and Gaze Estimation

Images of the eye are key in several computer vision problems, such as shape registration and gaze estimation. Recent large-scale supervised methods for these problems require time-consuming data collection and manual annotation, which can be unreliable. We propose synthesizing perfectly labelled photo-realistic training data in a fraction of the time. We used computer graphics techniques to build a collection of dynamic eye-region models from head scan geometry. These were randomly posed to synthesize close-up eye images for a wide range of head poses, gaze directions, and illumination conditions. We used our model's controllability to verify the importance of realistic illumination and shape variations in eye-region training data. Finally, we demonstrate the benefits of our synthesized training data (SynthesEyes) by out-performing state-of-the-art methods for eye-shape registration as well as cross-dataset appearance-based gaze estimation in the wild.

cs.CV↗

arXiv2015

Gracefully Degrading Consensus and $k$-Set Agreement in Directed Dynamic Networks

We study distributed agreement in synchronous directed dynamic networks, where an omniscient message adversary controls the availability of communication links. We prove that consensus is impossible under a message adversary that guarantees weak connectivity only, and introduce vertex-stable root components (VSRCs) as a means for circumventing this impossibility: A VSRC(k, d) message adversary guarantees that, eventually, there is an interval of $d$ consecutive rounds where every communication graph contains at most $k$ strongly (dynamic) connected components consisting of the same processes, which have at most outgoing links to the remaining processes. We present a consensus algorithm that works correctly under a VSRC(1, 4H + 2) message adversary, where $H$ is the dynamic causal network diameter. On the other hand, we show that consensus is impossible against a VSRC(1, H - 1) or a VSRC(2, $\infty$) message adversary, revealing that there is not much hope to deal with stronger message adversaries. However, we show that gracefully degrading consensus, which degrades to general $k$-set agreement in case of unfavourable network conditions, is feasible against stronger message adversaries: We provide a $k$-uniform $k$-set agreement algorithm, where the number of system-wide decision values $k$ is not encoded in the algorithm, but rather determined by the actual power of the message adversary in a run: Our algorithm guarantees at most $k$ decision values under a VSRC(n, d) + MAJINF(k) message adversary, which combines VSRC(n, d) (for some small $d$, ensuring termination) with some information flow guarantee MAJINF(k) between certain VSRCs (ensuring $k$-agreement). Our results provide a significant step towards the exact solvability/impossibility border of general $k$-set agreement in directed dynamic networks.

cs.DC↗

arXiv2014

Distributed Symmetry Breaking in Hypergraphs

Fundamental local symmetry breaking problems such as Maximal Independent Set (MIS) and coloring have been recognized as important by the community, and studied extensively in (standard) graphs. In particular, fast (i.e., logarithmic run time) randomized algorithms are well-established for MIS and $Δ+1$-coloring in both the LOCAL and CONGEST distributed computing models. On the other hand, comparatively much less is known on the complexity of distributed symmetry breaking in {\em hypergraphs}. In particular, a key question is whether a fast (randomized) algorithm for MIS exists for hypergraphs. In this paper, we study the distributed complexity of symmetry breaking in hypergraphs by presenting distributed randomized algorithms for a variety of fundamental problems under a natural distributed computing model for hypergraphs. We first show that MIS in hypergraphs (of arbitrary dimension) can be solved in $O(\log^2 n)$ rounds ($n$ is the number of nodes of the hypergraph) in the LOCAL model. We then present a key result of this paper --- an $O(Δ^ε\text{polylog}(n))$-round hypergraph MIS algorithm in the CONGEST model where $Δ$ is the maximum node degree of the hypergraph and $ε> 0$ is any arbitrarily small constant. To demonstrate the usefulness of hypergraph MIS, we present applications of our hypergraph algorithm to solving problems in (standard) graphs. In particular, the hypergraph MIS yields fast distributed algorithms for the {\em balanced minimal dominating set} problem (left open in Harris et al. [ICALP 2013]) and the {\em minimal connected dominating set problem}. We also present distributed algorithms for coloring, maximal matching, and maximal clique in hypergraphs.

cs.DC↗

arXiv2014

Distributed Agreement in Dynamic Peer-to-Peer Networks

Motivated by the need for robust and fast distributed computation in highly dynamic Peer-to-Peer (P2P) networks, we study algorithms for the fundamental distributed agreement problem. P2P networks are highly dynamic networks that experience heavy node {\em churn}. Our goal is to design fast algorithms (running in a small number of rounds) that guarantee, despite high node churn rate, that almost all nodes reach a stable agreement. Our main contributions are randomized distributed algorithms that guarantee {\em stable almost-everywhere agreement} with high probability even under high adversarial churn in a polylogarithmic number of rounds: 1. An $O(\log^2 n)$-round ($n$ is the stable network size) randomized algorithm that achieves almost-everywhere agreement with high probability under up to {\em linear} churn {\em per round} (i.e., $εn$, for some small constant $ε> 0$), assuming that the churn is controlled by an oblivious adversary (that has complete knowledge and control of what nodes join and leave and at what time and has unlimited computational power, but is oblivious to the random choices made by the algorithm). Our algorithm requires only polylogarithmic in $n$ bits to be processed and sent (per round) by each node. 2. An $O(\log m\log^3 n)$-round randomized algorithm that achieves almost-everywhere agreement with high probability under up to $ε\sqrt{n}$ churn per round (for some small $ε> 0$), where $m$ is the size of the input value domain, that works even under an adaptive adversary (that also knows the past random choices made by the algorithm). This algorithm requires up to polynomial in $n$ bits (and up to $O(\log m)$ bits) to be processed and sent (per round) by each node.

cs.DS↗

arXiv2014

The state of play of ASC-Inclusion: An Integrated Internet-Based Environment for Social Inclusion of Children with Autism Spectrum Conditions

Individuals with Autism Spectrum Conditions (ASC) have marked difficulties using verbal and non-verbal communication for social interaction. The running ASC-Inclusion project aims to help children with ASC by allowing them to learn how emotions can be expressed and recognised via playing games in a virtual world. The platform includes analysis of users' gestures, facial, and vocal expressions using standard microphone and web-cam or a depth sensor, training through games, text communication with peers, animation, video and audio clips. We present the state of play in realising such a serious game platform and provide results for the different modalities.

cs.HC↗

arXiv2013

DEX: Self-healing Expanders

We present a fully-distributed self-healing algorithm DEX, that maintains a constant degree expander network in a dynamic setting. To the best of our knowledge, our algorithm provides the first efficient distributed construction of expanders --- whose expansion properties hold {\em deterministically} --- that works even under an all-powerful adaptive adversary that controls the dynamic changes to the network (the adversary has unlimited computational power and knowledge of the entire network state, can decide which nodes join and leave and at what time, and knows the past random choices made by the algorithm). Previous distributed expander constructions typically provide only {\em probabilistic} guarantees on the network expansion which {\em rapidly degrade} in a dynamic setting; in particular, the expansion properties can degrade even more rapidly under {\em adversarial} insertions and deletions. Our algorithm provides efficient maintenance and incurs a low overhead per insertion/deletion by an adaptive adversary: only $O(\log n)$ rounds and $O(\log n)$ messages are needed with high probability ($n$ is the number of nodes currently in the network). The algorithm requires only a constant number of topology changes. Moreover, our algorithm allows for an efficient implementation and maintenance of a distributed hash table (DHT) on top of DEX, with only a constant additional overhead. Our results are a step towards implementing efficient self-healing networks that have \emph{guaranteed} properties (constant bounded degree and expansion) despite dynamic changes.

cs.DC↗

arXiv2013

Robust Leader Election in a Fast-Changing World

We consider the problem of electing a leader among nodes in a highly dynamic network where the adversary has unbounded capacity to insert and remove nodes (including the leader) from the network and change connectivity at will. We present a randomized Las Vegas algorithm that (re)elects a leader in O(D\log n) rounds with high probability, where D is a bound on the dynamic diameter of the network and n is the maximum number of nodes in the network at any point in time. We assume a model of broadcast-based communication where a node can send only 1 message of O(\log n) bits per round and is not aware of the receivers in advance. Thus, our results also apply to mobile wireless ad-hoc networks, improving over the optimal (for deterministic algorithms) O(Dn) solution presented at FOMC 2011. We show that our algorithm is optimal by proving that any randomized Las Vegas algorithm takes at least omega(D\log n) rounds to elect a leader with high probability, which shows that our algorithm yields the best possible (up to constants) termination time.

cs.DC↗

arXiv2013

Sublinear Bounds for Randomized Leader Election

This paper concerns {\em randomized} leader election in synchronous distributed networks. A distributed leader election algorithm is presented for complete $n$-node networks that runs in O(1) rounds and (with high probability) uses only $O(\sqrt{n}\log^{3/2} n)$ messages to elect a unique leader (with high probability). When considering the "explicit" variant of leader election where eventually every node knows the identity of the leader, our algorithm yields the asymptotically optimal bounds of O(1) rounds and O(n) messages. This algorithm is then extended to one solving leader election on any connected non-bipartite $n$-node graph $G$ in $O(τ(G))$ time and $O(τ(G)\sqrt{n}\log^{3/2} n)$ messages, where $τ(G)$ is the mixing time of a random walk on $G$. The above result implies highly efficient (sublinear running time and messages) leader election algorithms for networks with small mixing times, such as expanders and hypercubes. In contrast, previous leader election algorithms had at least linear message complexity even in complete graphs. Moreover, super-linear message lower bounds are known for time-efficient {\em deterministic} leader election algorithms. Finally, we present an almost matching lower bound for randomized leader election, showing that $Ω(\sqrt n)$ messages are needed for any leader election algorithm that succeeds with probability at least $1/e + \eps$, for any small constant $\eps > 0$. We view our results as a step towards understanding the randomized complexity ofleader election in distributed networks.

cs.DS↗

arXiv2013

Agreement in Directed Dynamic Networks

We study distributed computation in synchronous dynamic networks where an omniscient adversary controls the unidirectional communication links. Its behavior is modeled as a sequence of directed graphs representing the active (i.e. timely) communication links per round. We prove that consensus is impossible under some natural weak connectivity assumptions, and introduce vertex-stable root components as a means for circumventing this impossibility. Essentially, we assume that there is a short period of time during which an arbitrary part of the network remains strongly connected, while its interconnect topology may keep changing continuously. We present a consensus algorithm that works under this assumption, and prove its correctness. Our algorithm maintains a local estimate of the communication graphs, and applies techniques for detecting stable network properties and univalent system configurations. Our possibility results are complemented by several impossibility results and lower bounds for consensus and other distributed computing problems like leader election, revealing that our algorithm is asymptotically optimal.

cs.DC↗

arXiv2013

Storage and Search in Dynamic Peer-to-Peer Networks

We study robust and efficient distributed algorithms for searching, storing, and maintaining data in dynamic Peer-to-Peer (P2P) networks. P2P networks are highly dynamic networks that experience heavy node churn (i.e., nodes join and leave the network continuously over time). Our goal is to guarantee, despite high node churn rate, that a large number of nodes in the network can store, retrieve, and maintain a large number of data items. Our main contributions are fast randomized distributed algorithms that guarantee the above with high probability (whp) even under high adversarial churn: 1. A randomized distributed search algorithm that (whp) guarantees that searches from as many as $n - o(n)$ nodes ($n$ is the stable network size) succeed in ${O}(\log n)$-rounds despite ${O}(n/\log^{1+δ} n)$ churn, for any small constant $δ> 0$, per round. We assume that the churn is controlled by an oblivious adversary (that has complete knowledge and control of what nodes join and leave and at what time, but is oblivious to the random choices made by the algorithm). 2. A storage and maintenance algorithm that guarantees (whp) data items can be efficiently stored (with only $Θ(\log{n})$ copies of each data item) and maintained in a dynamic P2P network with churn rate up to ${O}(n/\log^{1+δ} n)$ per round. Our search algorithm together with our storage and maintenance algorithm guarantees that as many as $n - o(n)$ nodes can efficiently store, maintain, and search even under ${O}(n/\log^{1+δ} n)$ churn per round. Our algorithms require only polylogarithmic in $n$ bits to be processed and sent (per round) by each node. To the best of our knowledge, our algorithms are the first-known, fully-distributed storage and search algorithms that provably work under highly dynamic settings (i.e., high churn rates per step).

cs.DC↗

arXiv2011

Easy Impossibility Proofs for k-Set Agreement in Message Passing Systems

Despite of being quite similar agreement problems, consensus and general k-set agreement require surprisingly different techniques for proving the impossibility in asynchronous systems with crash failures: Rather than relatively simple bivalence arguments as in the impossibility proof for consensus (= 1-set agreement) in the presence of a single crash failure, known proofs for the impossibility of k-set agreement in systems with at least k>1 crash failures use algebraic topology or a variant of Sperner's Lemma. In this paper, we present a generic theorem for proving the impossibility of k-set agreement in various message passing settings, which is based on a simple reduction to the consensus impossibility in a certain subsystem. We demonstrate the broad applicability of our result by exploring the possibility/impossibility border of k-set agreement in several message-passing system models: (i) asynchronous systems with crash failures, (ii) partially synchronous processes with (initial) crash failures, and (iii) asynchronous systems augmented with failure detectors. In (i) and (ii), the impossibility part is just an instantiation of our main theorem, whereas the possibility of achieving k-set agreement in (ii) follows by generalizing the consensus algorithm for initial crashes by Fisher, Lynch and Patterson. In (iii), applying our technique yields the exact border for the parameter k where k-set agreement is solvable with the failure detector class (Sigma_k,Omega_k), for (1<= k<= n-1), of Bonnet and Raynal. Considering that Sigma_k was shown to be necessary for solving k-set agreement, this result yields new insights on the quest for the weakest failure detector.

cs.DC↗

arXiv2011

Solving k-Set Agreement with Stable Skeleton Graphs

In this paper we consider the k-set agreement problem in distributed message-passing systems using a round-based approach: Both synchrony of communication and failures are captured just by means of the messages that arrive within a round, resulting in round-by-round communication graphs that can be characterized by simple communication predicates. We introduce the weak communication predicate PSources(k) and show that it is tight for k-set agreement, in the following sense: We (i) prove that there is no algorithm for solving (k-1)-set agreement in systems characterized by PSources(k), and (ii) present a novel distributed algorithm that achieves k-set agreement in runs where PSources(k) holds. Our algorithm uses local approximations of the stable skeleton graph, which reflects the underlying perpetual synchrony of a run. We prove that this approximation is correct in all runs, regardless of the communication predicate, and show that graph-theoretic properties of the stable skeleton graph can be used to solve k-set agreement if PSources(k) holds.

cs.DC↗