LAMP

Category: Publications

  • Saliency for Fine-grained Object Recognition in Domains with Scarce Training Data

    Carola Figueroa Flores, Abel Gonzalez-García, Joost van de Weijer, Bogdan Raducanu Read Full Paper → This paper investigates the role of saliency to improve the classification accuracy of a Convolutional Neural Network (CNN) for the case when scarce training data is available. Our approach consists in adding a saliency branch to an existing CNN architecture which is used […]

  • Learning Metrics from Teachers: Compact Networks for Image Embedding

    Lu Yu, Vacit Oguz Yazici, Xialei Liu, Joost van de Weijer, Yongmei Cheng, Arnau Ramisa Read Full Paper → Metric learning networks are used to compute image embeddings, which are widely used in many applications such as image retrieval and face recognition. In this paper, we propose to use network distillation to efficiently compute image embeddings with small networks. Network […]

  • Learning effective RGB-D representations for scene recognition

    Xinhang Song, Shuqiang Jiang, Luis Herranz, Chengpeng Chen Read Full Paper → Deep convolutional networks can achieve impressive results on RGB scene recognition thanks to large data sets such as places. In contrast, RGB-D scene recognition is still underdeveloped in comparison, due to two limitations of RGB-D data we address in this paper. The first […]

  • Domain-adaptive deep network compression

    Marc Masana, Joost van de Weijer, Luis Herranz, Andrew D. Bagdanov, Jose M Alvarez Read Full Paper → Deep Neural Networks trained on large datasets can be easily transferred to new domains with far fewer labeled examples by a process called fine-tuning. This has the advantage that representations learned in the large source domain can be exploited on smaller […]

  • Rotate your Networks: Better Weight Consolidation and Less Catastrophic Forgetting

    Xialei Liu, Marc Masana, Luis Herranz, Joost Van de Weijer, Antonio M. Lopez, Andrew D. Bagdanov Read Full Paper → In this paper we propose an approach to avoiding catastrophic forgetting in sequential task learning scenarios. Our technique is based on a network reparameterization that approximately diagonalizes the Fisher Information Matrix of the network parameters. This reparameterization takes the form […]

  • Mix and match networks: encoder-decoder alignment for zero-pair image translation

    Yaxing Wang, Joost van de Weijer, Luis Herranz Read Full Paper → We address the problem of image translation between domains or modalities for which no direct paired data is available (i.e. zero-pair translation). We propose mix and match networks, based on multiple encoders and decoders aligned in such a way that other encoder-decoder pairs can be […]

  • Memory Replay GANs: learning to generate images from new categories without forgetting

    Chenshen Wu, Luis Herranz, Xialei Liu, Yaxing Wang, Joost van de Weijer, Bogdan Raducanu Read Full Paper → Previous works on sequential learning address the problem of forgetting in discriminative models. In this paper we consider the case of generative models. In particular, we investigate generative adversarial networks (GANs) in the task of learning new categories in a sequential fashion. […]

  • Exploiting Unlabeled Data in CNNs by Self-supervised Learning to Rank

    Xialei Liu, Joost van de Weijer, Andrew D. Bagdanov Read Full Paper → For many applications the collection of labeled data is expensive laborious. Exploitation of unlabeled data during training is thus a long pursued objective of machine learning. Self-supervised learning addresses this by positing an auxiliary task (different, but related to the supervised task) for which […]

  • Synthetic data generation for end-to-end thermal infrared tracking

    Lichao Zhang, Abel Gonzalez-Garcia, Joost van de Weijer, Martin Danelljan, Fahad Shahbaz Khan Read Full Paper → The usage of both off-the-shelf and end-to-end trained deep networks have significantly improved performance of visual tracking on RGB videos. However, the lack of large labeled datasets hampers the usage of convolutional neural networks for tracking in thermal infrared (TIR) images. Therefore, […]

  • Image-to-image translation for cross-domain disentanglement

    Abel Gonzalez-Garcia, Joost van de Weijer, Yoshua Bengio Read Full Paper → Deep image translation methods have recently shown excellent results, outputting high-quality images covering multiple modes of the data distribution. There has also been increased interest in disentangling the internal representations learned by deep methods to further improve their performance and achieve a finer control. In […]