Continual Learning for Visual Search with Backward Consistent Feature Embedding
This addresses the computational cost and compatibility issues in long-term visual search for applications like image retrieval, though it appears incremental as it builds on existing continual learning methods.
The paper tackles the problem of visual search with incrementally growing gallery sets by introducing a continual learning approach that maintains backward-consistent feature embeddings, eliminating the need to re-extract features for the entire gallery set. It shows efficacy across various benchmarks under diverse setups.
In visual search, the gallery set could be incrementally growing and added to the database in practice. However, existing methods rely on the model trained on the entire dataset, ignoring the continual updating of the model. Besides, as the model updates, the new model must re-extract features for the entire gallery set to maintain compatible feature space, imposing a high computational cost for a large gallery set. To address the issues of long-term visual search, we introduce a continual learning (CL) approach that can handle the incrementally growing gallery set with backward embedding consistency. We enforce the losses of inter-session data coherence, neighbor-session model coherence, and intra-session discrimination to conduct a continual learner. In addition to the disjoint setup, our CL solution also tackles the situation of increasingly adding new classes for the blurry boundary without assuming all categories known in the beginning and during model update. To our knowledge, this is the first CL method both tackling the issue of backward-consistent feature embedding and allowing novel classes to occur in the new sessions. Extensive experiments on various benchmarks show the efficacy of our approach under a wide range of setups.