Timezone: »
Poster
One Loss for All: Deep Hashing with a Single Cosine Similarity based Learning Objective
Jiun Tian Hoe · Kam Woh Ng · Tianyu Zhang · Chee Seng Chan · Yi-Zhe Song · Tao Xiang
A deep hashing model typically has two main learning objectives: to make the learned binary hash codes discriminative and to minimize a quantization error. With further constraints such as bit balance and code orthogonality, it is not uncommon for existing models to employ a large number (>4) of losses. This leads to difficulties in model training and subsequently impedes their effectiveness. In this work, we propose a novel deep hashing model with only $\textit{a single learning objective}$. Specifically, we show that maximizing the cosine similarity between the continuous codes and their corresponding $\textit{binary orthogonal codes}$ can ensure both hash code discriminativeness and quantization error minimization. Further, with this learning objective, code balancing can be achieved by simply using a Batch Normalization (BN) layer and multi-label classification is also straightforward with label smoothing. The result is a one-loss deep hashing model that removes all the hassles of tuning the weights of various losses. Importantly, extensive experiments show that our model is highly effective, outperforming the state-of-the-art multi-loss hashing models on three large-scale instance retrieval benchmarks, often by significant margins.
Author Information
Jiun Tian Hoe (Universiti Malaya)
Kam Woh Ng (Webank AI Lab)
Tianyu Zhang (GeekPlus)
Chee Seng Chan (Universiti Malaya)
Yi-Zhe Song
Tao Xiang (Samsung AI Centre, Cambridge)
More from the Same Authors
-
2021 Spotlight: SOFT: Softmax-free Transformer with Linear Complexity »
Jiachen Lu · Jinghan Yao · Junge Zhang · Xiatian Zhu · Hang Xu · Weiguo Gao · Chunjing XU · Tao Xiang · Li Zhang -
2023 Poster: HeadSculpt: Crafting 3D Head Avatars with Text »
Xiao Han · Yukang Cao · Kai Han · Xiatian Zhu · Jiankang Deng · Yi-Zhe Song · Tao Xiang · Kwan-Yee K. Wong -
2021 Poster: SOFT: Softmax-free Transformer with Linear Complexity »
Jiachen Lu · Jinghan Yao · Junge Zhang · Xiatian Zhu · Hang Xu · Weiguo Gao · Chunjing XU · Tao Xiang · Li Zhang -
2019 : Poster Session 2 »
Gabriele Prato · Urmish Thakker · Laura Galindez Olascoaga · Tianyu Zhang · Vahid Partovi Nia · Kamil Adamczewski -
2019 Poster: Rethinking Deep Neural Network Ownership Verification: Embedding Passports to Defeat Ambiguity Attacks »
Lixin Fan · Kam Woh Ng · Chee Seng Chan -
2018 Poster: Domain-Invariant Projection Learning for Zero-Shot Recognition »
An Zhao · Mingyu Ding · Jiechao Guan · Zhiwu Lu · Tao Xiang · Ji-Rong Wen