Bridging Non Co-occurrence with Unlabeled In-the-wild Data for Incremental Object Detection
Dec 6, 2021
Speakers
Yongqiang Zhang
Řečník · 0 sledujících
Mingli Ding
Řečník · 0 sledujících
Gim Hee Lee
Řečník · 0 sledujících
About
Deep networks have shown remarkable results in the task of object detection. However, their performance suffers critical drops when they are subsequently trained on novel classes without any sample from the base classes originally used to train the model. This phenomenon is known as catastrophic forgetting. Recently, several incremental learning methods are proposed to mitigate catastrophic forgetting for object detection. Despite the effectiveness, these methods require co-occurrence of the unlabeled base classes in the training data of the novel classes. This requirement is impractical in many real-world settings since the base classes do not necessarily co-occur with the novel classes. In view of this limitation, we consider a more practical setting of complete absence of co-occurrence of the base and novel classes for the object detection task. We propose the use of unlabeled in-the-wild data to bridge the non-occurrence caused by the missing base classes during the training of additional novel classes. To this end, we introduce a blind sampling strategy based on the responses of the base-class model and pre-trained novel-class model to select a smaller relevant dataset from the large in-the-wild dataset for incremental learning. We then design a dual-teacher distillation framework to transfer the knowledge distilled from the base- and novel-class teacher models to the student model using the sampled in-the-wild data. Experimental results on the PASCAL VOC and MS COCO datasets show that our proposed method significantly outperforms other state-of-the-art class-incremental object detection methods when there is no co-occurrence between the base and novel classes during training. Our code will be made open-source upon paper acceptance.Deep networks have shown remarkable results in the task of object detection. However, their performance suffers critical drops when they are subsequently trained on novel classes without any sample from the base classes originally used to train the model. This phenomenon is known as catastrophic forgetting. Recently, several incremental learning methods are proposed to mitigate catastrophic forgetting for object detection. Despite the effectiveness, these methods require co-occurrence of the unl…
Organizer
NeurIPS 2021
Účet · 1,9k sledujících
About NeurIPS 2021
Neural Information Processing Systems (NeurIPS) is a multi-track machine learning and computational neuroscience conference that includes invited talks, demonstrations, symposia and oral and poster presentations of refereed papers. Following the conference, there are workshops which provide a less formal setting.
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
PettingZoo: Gym for Multi-Agent Reinforcement Learning
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Benchmarks for Multi-Objective Hyperparameter Optimization
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Bootstrapping the Error of Oja's Algorithm
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Reward is enough for convex MDPs
Zhlédnout později
Tom Zahavy, …
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
General Low-rank Matrix Optimization: Geometric Analysis and Sharper Bounds
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Noisy Adaptation Generates Lévy Flights in Attractor Neural Networks
Zhlédnout později
Xingsi Dong, …
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %