TY - GEN
T1 - TopicFM
T2 - 37th AAAI Conference on Artificial Intelligence, AAAI 2023
AU - Giang, Khang Truong
AU - Song, Soohwan
AU - Jo, Sungho
N1 - Publisher Copyright:
Copyright © 2023, Association for the Advancement of Artificial Intelligence (www.aaai.org).
PY - 2023/6/27
Y1 - 2023/6/27
N2 - This study addresses an image-matching problem in challenging cases, such as large scene variations or textureless scenes. To gain robustness to such situations, most previous studies have attempted to encode the global contexts of a scene via graph neural networks or transformers. However, these contexts do not explicitly represent high-level contextual information, such as structural shapes or semantic instances; therefore, the encoded features are still not sufficiently discriminative in challenging scenes. We propose a novel image-matching method that applies a topic-modeling strategy to encode high-level contexts in images. The proposed method trains latent semantic instances called topics. It explicitly models an image as a multinomial distribution of topics, and then performs probabilistic feature matching. This approach improves the robustness of matching by focusing on the same semantic areas between the images. In addition, the inferred topics provide interpretability for matching the results, making our method explainable. Extensive experiments on outdoor and indoor datasets show that our method outperforms other state-of-the-art methods, particularly in challenging cases.
AB - This study addresses an image-matching problem in challenging cases, such as large scene variations or textureless scenes. To gain robustness to such situations, most previous studies have attempted to encode the global contexts of a scene via graph neural networks or transformers. However, these contexts do not explicitly represent high-level contextual information, such as structural shapes or semantic instances; therefore, the encoded features are still not sufficiently discriminative in challenging scenes. We propose a novel image-matching method that applies a topic-modeling strategy to encode high-level contexts in images. The proposed method trains latent semantic instances called topics. It explicitly models an image as a multinomial distribution of topics, and then performs probabilistic feature matching. This approach improves the robustness of matching by focusing on the same semantic areas between the images. In addition, the inferred topics provide interpretability for matching the results, making our method explainable. Extensive experiments on outdoor and indoor datasets show that our method outperforms other state-of-the-art methods, particularly in challenging cases.
UR - http://www.scopus.com/inward/record.url?scp=85165590267&partnerID=8YFLogxK
U2 - 10.1609/aaai.v37i2.25341
DO - 10.1609/aaai.v37i2.25341
M3 - Conference contribution
AN - SCOPUS:85165590267
T3 - Proceedings of the 37th AAAI Conference on Artificial Intelligence, AAAI 2023
SP - 2447
EP - 2455
BT - AAAI-23 Technical Tracks 2
A2 - Williams, Brian
A2 - Chen, Yiling
A2 - Neville, Jennifer
PB - AAAI Press
Y2 - 7 February 2023 through 14 February 2023
ER -