Inner Landscapes

Artist Statement

Inner Landscapes is a series of abstract dynamic paintings that explores emotion as a hidden structure unfolding across sound and image over time. Rather than depicting recognizable scenes or objects, each work invites viewers to experience emotion through evolving patterns of color, motion, and texture. As the paintings continuously grow and transform, viewers are encouraged to interpret their own emotional associations within these shifting visual landscapes.

Inspired by emotions such as calm, anxiety, nostalgia, grief, and joy, the series investigates how emotional meaning can emerge through cross-modal associations between music and visual form. Each piece begins with a reference image selected to represent an emotional state. Language model assistance recommends candidate images based on visual composition, atmosphere, and semantic associations, after which the artist selects the final reference image.

Each dynamic painting expresses a distinct emotional landscape. Calm unfolds through gentle movement and cool tones, anxiety through unstable branching and flickering intensity, nostalgia through fading and re-emerging traces, grief through dimming and dissolution, and joy through expanding forms and radiant color.

Mathematics, machine learning, computer vision, and stochastic processes function as artistic media. Through computational processes that translate between sound, image, motion, and meaning, the series reveals how neural networks can generate abstract yet recognizable emotional landscapes through cross-modal associations.

Technical Details

A coordinate-based neural network learns a continuous representation of the reference image from pixel coordinates while receiving the music spectrogram of the current time frame as input. The network is trained simultaneously across multiple zoom levels, allowing it to represent both broad compositional forms and fine-scale details. Different spectrogram thresholds emphasize different visual scales, causing the learned representation to continuously shift between larger and smaller structures. As the music evolves, the imagery appears to expand, contract, and breathe in response to changes in rhythm, intensity, and frequency.

To preserve visually significant features, an edge-based density map assigns greater training importance to regions with strong structural detail, while Fourier coordinate encoding enables the network to capture high-frequency image information. Together, these techniques create a balance between sharp, focused details and larger, softer painterly forms, producing a visual effect similar to selective focus.

The learned representation is rendered through a stochastic random-walk painting process. Strokes accumulate on a digital canvas while music intensity and emotion-specific styles influence their length, width, transparency, and directional randomness. The result is an evolving painting in which sound directly shapes motion, color, and structure.