Deep Manifold Traversal: Changing Labels with Convolutional Features
arXiv:1511.06421
Abstract
Many tasks in computer vision can be cast as a "label changing" problem, where the goal is to make a semantic change to the appearance of an image or some subject in an image in order to alter the class membership. Although successful task-specific methods have been developed for some label changing applications, to date no general purpose method exists. Motivated by this we propose deep manifold traversal, a method that addresses the problem in its most general form: it first approximates the manifold of natural images then morphs a test image along a traversal path away from a source class and towards a target class while staying near the manifold throughout. The resulting algorithm is surprisingly effective and versatile. It is completely data driven, requiring only an example set of images from the desired source and target domains. We demonstrate deep manifold traversal on highly diverse label changing tasks: changing an individual's appearance (age and hair color), changing the season of an outdoor image, and transforming a city skyline towards nighttime.
References in corpus (3)
Cited by in corpus (24)
- Densely Connected Convolutional Networks
- Detecting Adversarial Samples from Artifacts
- Deep Residual Correction Network for Partial Domain Adaptation
- Deep Photo Style Transfer
- Adversarial Examples in Modern Machine Learning: A Review
- GeneGAN: Learning Object Transfiguration and Attribute Subspace from Unpaired Data
- Model Patching: Closing the Subgroup Performance Gap with Data Augmentation
- Neural Face Editing with Intrinsic Image Disentangling
- ELEGANT: Exchanging Latent Encodings with GAN for Transferring Multiple Face Attributes
- Disentangled Representations for Manipulation of Sentiment in Text
- Instance-level Facial Attributes Transfer with Geometry-Aware Flow
- Discovering Fair Representations in the Data Domain
- A Factored Generalized Additive Model for Clinical Decision Support in the Operating Room
- Controllable Top-down Feature Transformer
- A Rotation Invariant Latent Factor Model for Moveme Discovery from Static Poses
- Learning to Succeed while Teaching to Fail: Privacy in Closed Machine Learning Systems
- Utility-aware Privacy-preserving Data Releasing
- On Kernel Method-Based Connectionist Models and Supervised Deep Learning Without Backpropagation
- Neural Crossbreed: Neural Based Image Metamorphosis
- Purifying Real Images with an Attention-guided Style Transfer Network for Gaze Estimation
- Analyzing Representations inside Convolutional Neural Networks
- Purifying Naturalistic Images through a Real-time Style Transfer Semantics Network
- Automatic Discoveries of Physical and Semantic Concepts via Association Priors of Neuron Groups
- Leveraging an Efficient and Semantic Location Embedding to Seek New Ports of Bike Share Services