Dynamic Gradient Reactivation for Backward Compatible Person Re-identification

by   Xiao Pan, et al.

We study the backward compatible problem for person re-identification (Re-ID), which aims to constrain the features of an updated new model to be comparable with the existing features from the old model in galleries. Most of the existing works adopt distillation-based methods, which focus on pushing new features to imitate the distribution of the old ones. However, the distillation-based methods are intrinsically sub-optimal since it forces the new feature space to imitate the inferior old feature space. To address this issue, we propose the Ranking-based Backward Compatible Learning (RBCL), which directly optimizes the ranking metric between new features and old features. Different from previous methods, RBCL only pushes the new features to find best-ranking positions in the old feature space instead of strictly alignment, and is in line with the ultimate goal of backward retrieval. However, the sharp sigmoid function used to make the ranking metric differentiable also incurs the gradient vanish issue, therefore stems the ranking refinement during the later period of training. To address this issue, we propose the Dynamic Gradient Reactivation (DGR), which can reactivate the suppressed gradients by adding dynamic computed constant during forward step. To further help targeting the best-ranking positions, we include the Neighbor Context Agents (NCAs) to approximate the entire old feature space during training. Unlike previous works which only test on the in-domain settings, we make the first attempt to introduce the cross-domain settings (including both supervised and unsupervised), which are more meaningful and difficult. The experimental results on all five settings show that the proposed RBCL outperforms previous state-of-the-art methods by large margins under all settings.


page 1

page 2

page 3

page 4


MixBCT: Towards Self-Adapting Backward-Compatible Training

The exponential growth of data, alongside advancements in model structur...

Neighborhood Consensus Contrastive Learning for Backward-Compatible Representation

In object re-identification (ReID), the development of deep learning tec...

Joint Disentangling and Adaptation for Cross-Domain Person Re-Identification

Although a significant progress has been witnessed in supervised person ...

FastFill: Efficient Compatible Model Update

In many retrieval systems the original high dimensional data (e.g., imag...

Privacy-Preserving Model Upgrades with Bidirectional Compatible Training in Image Retrieval

The task of privacy-preserving model upgrades in image retrieval desires...

BT^2: Backward-compatible Training with Basis Transformation

Modern retrieval system often requires recomputing the representation of...

Prediction with Unpredictable Feature Evolution

Feature space can change or evolve when learning with streaming data. Se...

Please sign up or login with your details

Forgot password? Click here to reset