# DenseImage Network: Video Spatial-Temporal Evolution Encoding and Understanding

Source: [DenseImage Network: Video Spatial-Temporal Evolution Encoding and Understanding](<https://www.sota2.com/research/paper/68fa645f1e209131cea0411d>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-04T02:59:32.365Z

Authors: Xiaokai Chen, Ke Gao

Published: 2018-05-19T08:53:44.000Z

## Abstract

Many of the leading approaches for video understanding are data-hungry and time-consuming, failing to capture the gist of spatial-temporal evolution in an efficient manner. The latest research shows that CNN network can reason about static relation of entities in images. To further exploit its capacity in dynamic evolution reasoning, we introduce a novel network module called DenseImage Network(DIN) with two main contributions. 1) A novel compact representation of video which distills its significant spatial-temporal evolution into a matrix called DenseImage, primed for efficient video encoding. 2) A simple yet powerful learning strategy based on DenseImage and a temporal-order-preserving CNN network is proposed for video understanding, which contains a local temporal correlation constraint capturing temporal evolution at multiple time scales with different filter widths. Extensive experiments on two recent challenging benchmarks demonstrate that our DenseImage Network can accurately capture the common spatial-temporal evolution between similar actions, even with enormous visual variations or different time scales. Moreover, we obtain the state-of-the-art results in action and gesture recognition with much less time-and-memory cost, indicating its immense potential in video representing and understanding.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Action Recognition on Something-Something (test)](<https://www.sota2.com/research/sota/action-recognition-on-something-something-test.md>): Action Recognition; Something-Something (test); Top-1 Acc: 34.11.

- [Hand Gesture Recognition on Jester 1 (test)](<https://www.sota2.com/research/sota/hand-gesture-recognition-on-jester-1-test.md>): Hand Gesture Recognition; Jester 1 (test); Top-1 Accuracy: 95.31.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1805.07550>)

[PDF](<https://arxiv.org/pdf/1805.07550.pdf>)