TokenLearner: What Can 8 Learned Tokens Do for Images and Videos?
read the original abstract
In this paper, we introduce a novel visual representation learning which relies on a handful of adaptively learned tokens, and which is applicable to both image and video understanding tasks. Instead of relying on hand-designed splitting strategies to obtain visual tokens and processing a large number of densely sampled patches for attention, our approach learns to mine important tokens in visual data. This results in efficiently and effectively finding a few important visual tokens and enables modeling of pairwise attention between such tokens, over a longer temporal horizon for videos, or the spatial content in images. Our experiments demonstrate strong performance on several challenging benchmarks for both image and video recognition tasks. Importantly, due to our tokens being adaptive, we accomplish competitive results at significantly reduced compute amount. We obtain comparable results to the state-of-the-arts on ImageNet while being computationally more efficient. We also confirm the effectiveness of the approach on multiple video datasets, including Kinetics-400, Kinetics-600, Charades, and AViD. The code is available at: https://github.com/google-research/scenic/tree/main/scenic/projects/token_learner
This paper has not been read by Pith yet.
Forward citations
Cited by 11 Pith papers
-
EgoSAT: A Comprehensive Benchmark of Egocentric Streaming Interaction Understanding
EgoSAT is the first benchmark unifying retrospective, online, and prospective reasoning tasks in egocentric streaming video to evaluate VLMs, revealing struggles with temporal modeling and mis-calibration.
-
Why Training-Free Token Reduction Collapses: The Inherent Instability of Pairwise Scoring Signals
Pairwise scoring signals in Vision Transformer token reduction are inherently unstable due to high perturbation counts and degrade in deep layers, causing collapse, while unary signals with triage enable CATIS to reta...
-
Spatial-Aware Reduction Framework: Towards Efficient and Faithful Visual State Space Models
STORM is a training-free spatial-aware token reduction framework that reformulates compression on spatial units to preserve grid topology and neighborhood coherence in visual state space models.
-
See Less, Specify More: Visual Evidence Budgets for Generalizable VLAs
S2 improves generalization in vision-language-action models by using goal-preserving refined language guidance and explicit visual evidence budgets, raising mean subtask success from 54.2% to 79.0% on eight real-robot...
-
FlowNar: Scalable Streaming Narration for Long-Form Videos
FlowNar achieves bounded memory and 3x higher throughput for streaming narration on Ego4D, EgoExo4D, and EpicKitchens100 by combining dynamic historical context removal with a Cross Linear Attentive Memory module.
-
Head Similarity: Modeling Structured Whole-Head Appearance Beyond Face Recognition
Head Similarity extends identity recognition to structured whole-head similarity by capturing intra-identity appearance variations via hierarchical supervision on a weakly-labeled video benchmark.
-
One Trajectory, One Token: Grounded Video Tokenization via Panoptic Sub-object Trajectory
TrajViT tokenizes videos via panoptic sub-object trajectories, achieving 10x token reduction and outperforming ViT3D by 6% on retrieval and 5.2% on VideoQA tasks with faster training and inference.
-
PaLM-E: An Embodied Multimodal Language Model
PaLM-E is a single 562B-parameter multimodal model that performs embodied reasoning tasks like robotic manipulation planning and visual question answering by interleaving vision, state, and text inputs with positive t...
-
Florence: A New Foundation Model for Computer Vision
Florence is a new vision foundation model that learns universal visual-language representations from web-scale data and reports state-of-the-art results on 44 benchmarks including 83.74% zero-shot ImageNet top-1 accuracy.
-
DinoLink: A Token-Centric Representation Compression Framework for Bandwidth-Constrained Collaborative V2X Perception
DinoLink uses saliency-aware token pruning and residual vector quantization to cut V2X bitrate by 139x while retaining 32.8% mAP on nuScenes.
-
DinoLink: A Token-Centric Representation Compression Framework for Bandwidth-Constrained Collaborative V2X Perception
DinoLink uses saliency-aware token pruning plus residual vector quantization to cut V2X bitrate by 139x while reporting 32.8% mAP on nuScenes.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.