DeepAI AI Chat
Log In Sign Up

Perceiver: General Perception with Iterative Attention

by   Andrew Jaegle, et al.

Biological systems understand the world by simultaneously processing high-dimensional inputs from modalities as diverse as vision, audition, touch, proprioception, etc. The perception models used in deep learning on the other hand are designed for individual modalities, often relying on domain-specific assumptions such as the local grid structures exploited by virtually all existing vision models. These priors introduce helpful inductive biases, but also lock models to individual modalities. In this paper we introduce the Perceiver - a model that builds upon Transformers and hence makes few architectural assumptions about the relationship between its inputs, but that also scales to hundreds of thousands of inputs, like ConvNets. The model leverages an asymmetric attention mechanism to iteratively distill inputs into a tight latent bottleneck, allowing it to scale to handle very large inputs. We show that this architecture performs competitively or beyond strong, specialized models on classification tasks across various modalities: images, point clouds, audio, video and video+audio. The Perceiver obtains performance comparable to ResNet-50 on ImageNet without convolutions and by directly attending to 50,000 pixels. It also surpasses state-of-the-art results for all modalities in AudioSet.


page 3

page 6

page 12

page 15

page 16


Attention Bottlenecks for Multimodal Fusion

Humans perceive the world by concurrently processing and fusing high-dim...

Hierarchical Perceiver

General perception systems such as Perceivers can process arbitrary moda...

Input-level Inductive Biases for 3D Reconstruction

Much of the recent progress in 3D vision has been driven by the developm...

PolyViT: Co-training Vision Transformers on Images, Videos and Audio

Can we train a single transformer model capable of processing multiple m...

Zorro: the masked multimodal transformer

Attention-based models are appealing for multimodal processing because i...

Hypernetworks build Implicit Neural Representations of Sounds

Implicit Neural Representations (INRs) are nowadays used to represent mu...

Scalable Adaptive Computation for Iterative Generation

We present the Recurrent Interface Network (RIN), a neural net architect...

Code Repositories


Implementation of Perceiver, General Perception with Iterative Attention, in Pytorch

view repo


Implementation of Perceiver, General Perception with Iterative Attention in TensorFlow

view repo


Perceiver (transformer variant) implemented in JAX and Flax

view repo


Just a quick toy example of the Perceiver architecture (PyTorch & PyTorch Lightning)

view repo