在处理基于骨骼点的连续姿态动作识别时，Transformer与LSTM在适用性上各有特点。 LSTM是RNN的变体，在处理视频序列数据时，能利用时间信息提高姿态估计的准确性，对于动态姿态的连续性估计较好。不过，由于其按顺序处理数据的特性，计算效率相对较低，在捕捉长距离依赖关系方面也存在一定局限，对于复杂姿态动作中骨骼点之间的长距离依赖关系的捕捉能力不如Transformer [^1]。 Transformer凭借自注意力机制，能够有效捕捉长距离依赖关系，在复杂姿态的准确性估计上表现突出，尤其是在处理大规模、多样化的数据集时，能够学习到更全面的姿态特征，从而提高准确性。在基于骨骼点的连续姿态动作识别中，若动作复杂，骨骼点之间存在长距离依赖关系，Transformer会更具优势 [^1]。 以下是一个简单的LSTM和Transformer模型在Keras中的示例代码： ```python import tensorflow as tf from tensorflow.keras import layers # LSTM模型示例 lstm_model = tf.keras.Sequential([ layers.LSTM(64, input_shape=(None, 10)), layers.Dense(1, activation='sigmoid') ]) # Transformer模型简单示例（简化版） class TransformerBlock(layers.Layer): def __init__(self, embed_dim, num_heads, ff_dim, rate=0.1): super(TransformerBlock, self).__init__() self.att = layers.MultiHeadAttention(num_heads=num_heads, key_dim=embed_dim) self.ffn = tf.keras.Sequential( [layers.Dense(ff_dim, activation="relu"), layers.Dense(embed_dim),] ) self.layernorm1 = layers.LayerNormalization(epsilon=1e-6) self.layernorm2 = layers.LayerNormalization(epsilon=1e-6) self.dropout1 = layers.Dropout(rate) self.dropout2 = layers.Dropout(rate) def call(self, inputs, training): attn_output = self.att(inputs, inputs) attn_output = self.dropout1(attn_output, training=training) out1 = self.layernorm1(inputs + attn_output) ffn_output = self.ffn(out1) ffn_output = self.dropout2(ffn_output, training=training) return self.layernorm2(out1 + ffn_output) embed_dim = 32 # Embedding size for each token num_heads = 2 # Number of attention heads ff_dim = 32 # Hidden layer size in feed forward network inside transformer inputs = layers.Input(shape=(None, 10)) transformer_block = TransformerBlock(embed_dim, num_heads, ff_dim) x = transformer_block(inputs) x = layers.GlobalAveragePooling1D()(x) x = layers.Dropout(0.1)(x) x = layers.Dense(20, activation="relu")(x) x = layers.Dropout(0.1)(x) outputs = layers.Dense(1, activation="sigmoid")(x) transformer_model = tf.keras.Model(inputs=inputs, outputs=outputs) ```