arXiv subjects

Hoang Le

Publications and source records attributed to Hoang Le.

22 records · Page 2Linked to original sources

arXiv2020

Deep Homography Estimation for Dynamic Scenes

Homography estimation is an important step in many computer vision problems. Recently, deep neural network methods have shown to be favorable for this problem when compared to traditional methods. However, these new methods do not consider dynamic content in input images. They train neural networks with only image pairs that can be perfectly aligned using homographies. This paper investigates and discusses how to design and train a deep neural network that handles dynamic scenes. We first collect a large video dataset with dynamic content. We then develop a multi-scale neural network and show that when properly trained using our new dataset, this neural network can already handle dynamic scenes to some extent. To estimate a homography of a dynamic scene in a more principled way, we need to identify the dynamic content. Since dynamic content detection and homography estimation are two tightly coupled tasks, we follow the multi-task learning principles and augment our multi-scale network such that it jointly estimates the dynamics masks and homographies. Our experiments show that our method can robustly estimate homography for challenging scenarios with dynamic scenes, blur artifacts, or lack of textures.

cs.CV↗

arXiv2018

Tropical Vertex-Disjoint Cycles of a Vertex-Colored Digraph: Barter Exchange with Multiple Items Per Agent

In a barter exchange market, agents bring items and seek to exchange their items with one another. Agents may agree to a k-way exchange involving a cycle of k agents. A barter exchange market can be represented by a digraph where the vertices represent items and the edges out of a vertex indicate the items that an agent is willing to accept in exchange for that item. It is known that the problem of finding a set of vertex-disjoint cycles with the maximum total number of vertices (MAX-SIZE-EXCHANGE) can be solved in polynomial time. We consider a barter exchange where each agent may bring multiple items, and items of the same agent are represented by vertices with the same color. A set of cycles is said to be tropical if for every color there is a cycle that contains a vertex of that color. We show that the problem of determining whether there exists a tropical set of vertex-disjoint cycles in a digraph (TROPICAL-EXCHANGE) is NP-complete and APX-hard. This is equivalent to determining whether it is possible to arrange an exchange of items among agents such that every agent trades away at least one item. TROPICAL-MAX-SIZE-EXCHANGE is a similar problem, where the goal is to find a set of vertex-disjoint cycles that contains the maximum number of vertices and also contains all of the colors in the graph. We show that this problem is likewise NP-complete and APX-hard. For the restricted case where there are at most two vertices of each color (corresponding to a restriction that each agent may bring at most two items), both problems remain NP-hard but are in APX. Finally, we consider MAX-SIZE-TROPICAL-EXCHANGE, where the set of cycles must primarily include as many colors as possible and secondarily include as many vertices as possible. We show that this problem is NP-hard.

cs.DS↗

arXiv2017

Scaling Properties of Human Brain Functional Networks

We investigate scaling properties of human brain functional networks in the resting-state. Analyzing network degree distributions, we statistically test whether their tails scale as power-law or not. Initial studies, based on least-squares fitting, were shown to be inadequate for precise estimation of power-law distributions. Subsequently, methods based on maximum-likelihood estimators have been proposed and applied to address this question. Nevertheless, no clear consensus has emerged, mainly because results have shown substantial variability depending on the data-set used or its resolution. In this study, we work with high-resolution data (10K nodes) from the Human Connectome Project and take into account network weights. We test for the power-law, exponential, log-normal and generalized Pareto distributions. Our results show that the statistics generally do not support a power-law, but instead these degree distributions tend towards the thin-tail limit of the generalized Pareto model. This may have implications for the number of hubs in human brain functional networks.

q-bio.NC↗

arXiv2011

Bidirectional Pipelining for Scalable IP Lookup and Packet Classification

Both IP lookup and packet classification in IP routers can be implemented by some form of tree traversal. SRAM-based Pipelining can improve the throughput dramatically. However, previous pipelining schemes result in unbalanced memory allocation over the pipeline stages. This has been identified as a major challenge for scalable pipelined solutions. This paper proposes a flexible bidirectional linear pipeline architecture based on widely-used dual-port SRAMs. A search tree is partitioned, and then mapped onto pipeline stages by a bidirectional fine-grained mapping scheme. We introduce the notion of inversion factor and several heuristics to invert subtrees for memory balancing. Due to its linear structure, the architecture maintains packet input order, and supports non-blocking route updates. Our experiments show that, the architecture can achieve a perfectly balanced memory distribution over the pipeline stages, for both trie-based IP lookup and tree-based multi-dimensional packet classification. For IP lookup, it can store a full backbone routing table with 154419 entries using 2MB of memory, and sustain a high throughput of 1.87 billion packets per second (GPPS), i.e. 0.6 Tbps for the minimum size (40 bytes) packets. The throughput can be improved further to be 2.4 Tbps, by employing caching to exploit the Internet traffic locality.

cs.NI↗