Author name cluster

Weiguo Fan

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

4 papers

2 author rows

ECAI Conference 2020 Conference Paper

Learning Class-Based Graph Representation for Object Detection

Shuyu Miao
Rui Feng 0001
Yuejie Zhang
Weiguo Fan

Object detection has achieved a tremendous advancement based on feature-based learning in the vision space, while little work has focused on reasoning in the perception space like humans. One of the greatest challenges lies in that it is difficult to build a connectivity model in the topological space for relational reasoning, since the current network is better at modeling the distribution of structured data. To settle this issue, we introduce a novel graph modeling mechanism with class-based graph representation, which contributes to modeling the high-order topology structure that maps the data distribution to make the detection models have better relational reasoning ability. In this mechanism, we propose three learning subtasks, i. e. , vision-to-perception embedding, perception reasoning graph representation, and perception-to-vision modeling. The mechanism based on such subtasks effectively maintains the independence of the original detection network and the proposed mechanism-based model, thus it can be well integrated with existing detection models without additional modification. The experimental results demonstrate the feasibility and effectiveness of our proposed mechanism, and the new state-of-the-art performance can be achieved on the public challenging datasets for object detection.

Details

AAAI Conference 2020 Conference Paper

Zero-Shot Sketch-Based Image Retrieval via Graph Convolution Network

Zhaolong Zhang
Yuejie Zhang
Rui Feng
Tao Zhang
Weiguo Fan

Zero-Shot Sketch-based Image Retrieval (ZS-SBIR) has been proposed recently, putting the traditional Sketch-based Image Retrieval (SBIR) under the setting of zero-shot learning. Dealing with both the challenges in SBIR and zero-shot learning makes it become a more difﬁcult task. Previous works mainly focus on utilizing one kind of information, i. e. , the visual information or the semantic information. In this paper, we propose a SketchGCN model utilizing the graph convolution network, which simultaneously considers both the visual information and the semantic information. Thus, our model can effectively narrow the domain gap and transfer the knowledge. Furthermore, we generate the semantic information from the visual information using a Conditional Variational Autoencoder rather than only map them back from the visual space to the semantic space, which enhances the generalization ability of our model. Besides, feature loss, classi- ﬁcation loss, and semantic loss are introduced to optimize our proposed SketchGCN model. Our model gets a good performance on the challenging Sketchy and TU-Berlin datasets.

PDF Details

AAAI Conference 2019 Conference Paper

Fully Convolutional Video Captioning with Coarse-to-Fine and Inherited Attention

Kuncheng Fang
Lian Zhou
Cheng Jin
Yuejie Zhang
Kangnian Weng
Tao Zhang
Weiguo Fan

Automatically generating natural language description for video is an extremely complicated and challenging task. To tackle the obstacles of traditional LSTM-based model for video captioning, we propose a novel architecture to generate the optimal descriptions for videos, which focuses on constructing a new network structure that can generate sentences superior to the basic model with LSTM, and establishing special attention mechanisms that can provide more useful visual information for caption generation. This scheme discards the traditional LSTM, and exploits the fully convolutional network with coarse-to-fine and inherited attention designed according to the characteristics of fully convolutional structure. Our model cannot only outperform the basic LSTM-based model, but also achieve the comparable performance with those of state-of-the-art methods.

PDF Details

IS Journal 2006 Journal Article

Adaptive Web Search: Evolving a Program That Finds Information

Gordon
Weiguo Fan
Pathak

Search engines contain programs that compare the words in a user's query to the words and phrases in Web pages. This comparison emphasizes relatively rare terms, terms that occur frequently in a page, and terms in prominent positions (such as a page's title), among other textual clues that suggest what the page is about. Although all search engines differ in the ways they determine which Web pages to present to a user, each incorporates a method that its designers hope will be effective. Nonetheless, retrieval algorithms perform inconsistently—some better in one circumstance, others in another--with no way to know in advance which will be most effective. The authors approach retrieval from a learning perspective. Rather than determining how to combine lexical clues beforehand, they infer how this should be done on the basis of users' evaluations of previously viewed documents. Unlike conventional systems, this approach automatically evolves new retrieval programs through genetic programming. It seems particularly effective for users whose need for information remains consistent over weeks or months.

Details DOI